[odf-discuss] Response from a GNOME Foundation board director

Pamela Jones pj at groklaw.net
Fri Nov 2 05:45:21 EDT 2007


Please allow me to build on what Ian is saying. It may help to show an 
explicit example of how Microsoft traditionally approaches what Jody is 
approaching from a technical perspective only.

This is an exhibit made public in the Comes v. Microsoft litigation:

Here's the complete "Generalized Evangelism Timeline" document,
with html tags. The first page was actually in landscape format, so I
just typed it as an ordinary page.
<html><body></body></html> tags were added so Konqueror could render
it :-)


Regards,
Bjørn/kattemann







(page 1 of 27 is blank except for footer)

Generalized Evangelism Timeline

    1. Evangelism starts. Goals, strategy & tactics, ISVs, and 
influencers, are defined. Verify that spec & demos are nearing 
stability. Verify PSS & dev team commitment to support ISVs in DevLab. 
Identify target ISVs, and the potential carrots & sticks that you can 
use to motivate them. Think about development tool issues (most other 
ISVs will need tools to implement support for your technology; what 
products are critical? Target them first.)
    2. DR Prep.Preparations for a Design Review (DR) begin. Line up 
speakers, logistics, invitations, skeleton press release, "showcase" 
ISVs, etc.
    3. Host DR. Upon availability of (a) a spec (S1) and/or (b) first 
internal build stable enough to demo, host the targeted ISVs at a Design 
Review. Goals: (i) verify that the technology is useful to ISVs,(ii)get 
ISVs emotionally involved in the technology, (iii) generate positive 
press regarding the technology.
    4. Limited Developers' Release. Technology & specification reach 
sufficient stability (A1) to be released to the targeted developers. 
(Stages 3 & 4 are often collapsed together, such that the first stable 
release is given to the Design Review attendees.) This is not a general 
beta release.
    5. Jihad At (or soon after) the Design Review, send ISVs the details 
of the [Technology Name] FirstWave Program, including the Letter of 
Agreement, for their review. Within a few days, visit each targeted ISV 
in person, to close their participation in the FirstWave Program.
    6. FirstWave Program. Host the targeted ISVs in a series of visits 
(rarely more than three) in the DevLab, focused on producing versions of 
their apps that demonstrate the ISV's support for [Technology Name], 
Monitor progress of co-marketing benefits. Monitor progress of ISVs 
towards successful demonstration; send details up the management chain. 
Consolidate ISV feedback, bug fix requests, etc., and generally 
represent the ISVs to the technology product team. Ensure that adequate 
sample code and documentation is being produced. Line up courseware, 
books (MS Press and third party), articles, conference sessions, stacked 
panel discussions, seminars, maganzine articles, defections from the 
competition, rifts in alliances, etc.
    7. Beta Release Upon the wide release of a stable beta of the 
technology to large numbers of developers (e.g. via MSDN or the Web), 
host an event for ISVs, press, analysts, allies and competitors, at 
which the targeted ISVs demonstrate their apps' support of the 
technology, as produced in the DevLab.
    8. The Slog. Work with early adopters to help get their products 
ready to shi. Represent the ISVs at bug-prioritizing meetings. Speak 
with press, analysts, ISVs, allies, and competitors. Broaden evangelism 
efforts through the involvement of MSDN's manifold resources. Execute 
the tactics ar previously determined, with minor modifications on the 
fly: this is the battle phase, win or lose.
    9. Final Release. Technology is released in final form. Big event 
for press, analysts, ISVs, allies, competitors, etc., showing momentum – 
early adopters demo apps (which are hopefully at or near final 
themselves), plus demos of other cool apps that turned up during the 
Slog. The Slog may continue for some time.
   10. Critical Mass. The technology is recognized as de facto standard, 
and the vaast majority of developers incorporate it into their future 
plans. Some influencers may actually increase their support for 
competing technologies, but it's too late   critical mass has been 
achieved, and they are at Ground Zero.
   11. Mopping Up. During the mopping-up phase, ensure that the enemy 
technology is routed. Use the press, the Internet, etc. to heighten the 
impression that the enemy is desperate, demoralized, defeated, 
depressed. Usually, this phase (or even Phase 8, the Slog) overlaps 
phases 1-3 of the next version of [Technology Name], which addresses all 
of the advantages of the competitors' technology, while addressing 
[Technology Name]'s key weaknesses. Repeat phases 1-10 as necessary 
/e.g. Windows NT 3.1 through 4.0; OLE 1.0 through COM+; Windows 1.0 
through Windows 3.1).
   12. Victory. The developers, marketers, and managers of the competing 
technology give up the sinking ship, and interview for positions at 
Microsoft.

       MS-PCA 1913135
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 2 of 27

       Generalized Evangelism Timeline

       James Plamondon
       Technical Evangelist
       Microsoft Developer Relations Group
       October 9, 1997

       Details

       1: Evangelism Starts

       Work closely with the technology development team, to ensure that 
they are developing the right technology, the right way. The right 
technology is that which will make money for Microsoft, or prevent 
Microsoft from losing money, whether directly or indirectly. The right 
way to develop technology is with a clear vision of how the technology 
is going to meet that objective – with a clear, concise focus on what 
actions need to be taken – by whom, when, how, and for what reason – in 
order to achieve that objective.

       One factor – but by no means the only factor – in verifying that 
the technology being developed is the right one, is ensuring that the 
technology meets the needs of any ISVs that we will need to have support 
it. If the technology can succeed without third-party support, then this 
is not a consideration, and Evangelism will be relatively uninvolved in 
the process. On the other hand, if ISV support is a key factor in the 
success of the technology, then Evangelism's role will be critical.

       Well before there is a detailed spec, decide (with the technology 
team) which ISVs' support is (a) most critical to the success of the 
technology, and (b) most easily gained, for whatever reasons. Under NDA, 
and with the support of the technology team, discuss the technology 
informally with a small number (four, plus or minus two) of these 
critical ISVs, to get a "reality check" on ISV reactions to the 
technology. Make sure that the feedback from these ISVs is either 
incorporated into the spec, or discarded with good reason. If the 
feedback is discarded, formulate compelling Q&A explaining the reasons 
for discarding it. Run the results past your small number of ISVs, to 
verify that it is compelling. If not, rethink the discarded feedback. If 
the overall feedback is negative, rethink development of the technology.

       Now, you are prepared to define the evangelism plan. First, 
define the goal of the technology. This is usually quite simple: to 
become the de facto industry standard mechanism for [accomplishing some 
task], To achieve this overall goal, one must achieve a number of 
specific, measurable objectives along the way. Some examples of 
objectives might be
           * Having n tool vendors commit to shipping support for 
[technology name] in their tools by [some specific date]
           * Having m ISVs demonstrate support for [technology name] in 
beta versions of their apps at [some specific venue]
           * Having y books on [technology name] be available by [date], 
to help developers implement support for [technology name] in their 
applications

       Presuming that ISV support is important to the success of the 
technology, then there usually exists an essentially infinite number of 
ISVs that could support the technology. However, Evangelism usually 
lacks the resources necessary to work closely with all of these ISVs. 
Further, some of these ISVs will contribute much more to the success or 
failure of the technology than others. Therefore, Evangelism usually 
must select a subset of ISVs to work with, out of the complete set of 
ISVs with whom it could work.

       Those ISVs should be selected for evangelism, which will deliver 
the most benefit for the least amount of work. Some will be selected 
because they are easy wins - a company that has expressly and joyfully 
tied its success very closely to our own, or a company which is run by 
an ex-Softie who misses working with Microsoft. Some will be selected 
because they are the most influential, either in terms of market share or

       MS-PCA 1913137
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 3 of 27

       mind share. (Mind share denotes the extent to which a product is 
discussed; market share denotes the extent to which a product is used.)

       The ISVs targeted for evangelism should be categorized into four 
tiers:
          1. Tier A: The most influential ISVs, in terms of market share 
or mind share. These will usually be obvious, but always do some 
research anyway to make sure that you're not overlooking anyone,
          2. Tier B: Less influential ISVs, who might be worth 
evangelizing anyway – perhaps because they are
                1. Easy wins
                2. Generally assumed to be in the enemy camp, such that 
their supporting your technology would be a shock (and, hopefully cause 
a rift in the enemy camp)
                3. Utterly worthless evangelically, but you can lead the 
enemy to waste a lot of resources evangelizing them, if the enemy knows 
that "Microsoft cares a lot" about them.
          3. Tier C: ISVs utterly lacking in influence, but whom you 
might speak to about the technology, if you can reach them in a 
one-to-many fashion – perhaps while you're visiting Tier A accounts in 
their city.
          4. Tier Z: ISVs that should never know that you exist, lest 
they send you email, call you, or otherwise waste your time.

       For some technologies, Independent Content Providers (ICPs) will 
be more relevant than Independent Software Providers (ISVs). For some, 
you may have to involve hardware vendors, advertising agencies, or other 
providers of evangelical leverage. I use "ISV" in this paper to 
generally denote all of these kinds of evangelism targets; make whatever 
substitutions are necessary for your specific technology, service, or 
whatever.

       In addition to identifying and categorizing the relevant ISVs, 
Evangelism should also identify and categorize other industry 
influencers during this first phase of evangelism. There are three 
categories of industry influencers:
          1. Providers of evangelical infrastructure: The authors of 
technical books, courseware designers and instructors, authors of 
technical articles, conference organizers, software consultants and 
contract development houses &emdash; Evangelism needs to involve all of 
these folks at various stages of an evangelism campaign, to make sure 
that ISVs and individual developers have the information they need to 
(a) decide to support the technology, and to (b) implement the 
technology in their application.
          2. The Press: Almost every evangelism campaign involves 
working with the press, either directly or through a PR agency. Our 
evangelism plan should identify the specific members of the press that 
you will target for technical evangelism (as distinct from the usual, 
non-technical PR treatment).
          3. Analysts: Analysts are people who are paid to take a stand, 
while always trying to appear to be disinterested observers (since the 
appearance of independence maximizes the price they can charge for 
selling out). Treat them as you would treat nuclear weapons – as an 
important part of your arsenal, which you want to keep out of the hands 
of the enemy. Bribe Hire them to produce "studies" that "prove" that 
your technology is superior to the enemy's, and that it is gaining 
momentum faster.

       Having identified the objectives sought, the ISVs whose support 
is necessary to achieve that support, and the specific carrots and 
sticks that you are going to use to gain their support, you are prepared 
to document your proposed plan in an Evangelism Plan. A sample 
Evangelism Plan can be found at the end of this document [1]

       After the completion of the Plan, you are ready to prepare for 
the presentation of the technology to a group of ISVs: the Design Review,

       2: Design Review Preparation
       A Design Review is a gathering of evangelism targets, usually on 
campus, to whom Microsoft presents an early version of the technology. 
The goals of a Design Review are four-fold:

       [1] Thanks to Peter Plamondon.

       MS-PCA 1913138
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 4 of 27

          1. To verify that the technology is genuinely useful to ISVs, 
without any glaring weaknesses that will prevent its being used in the 
intended manner –- and to gather any suggestions that might improve it
          2. To make the attendees feel honored, respected, and 
involved, such that they become emotionally attached to the technology, 
and are therefore more likely to support it in their products
          3. To act as a forcing-function for the technology team, by 
requiring them to produce clear descriptions, demos, and specs.
          4. To generate positive press regarding the technology

       To ensure that these objectives are accomplished at the Design 
Review, Evangelism must prepare extensively before the DR takes place. A 
date must be selected, to which the technology team will commit. Meeting 
room space must be reserved (usually in Building 12, or at a nearby 
hotel's conference center) and other logistics (meals, handouts, etc) 
arranged. A draft press release must be approved, pending final quotes 
from the attendees (gathered at the end of the DR). Speakers must be 
arranged, their slides reviewed, their demos tested, etc. – and the 
evangelist owns the whole shebang. The evangelist doesn't have to do it 
all, and indeed should do as little of it as possible. But the 
evangelist has to ensure that it all gets done properly, and on time.

       Before the DR, various steps can be taken behind the scenes to 
ensure that the result is positive. Many of these steps are right out of 
books such as "What They Don't Teach You At The Harward Busines School," 
or manuals on winning at office politics, or the like. For example, you 
should always know what your ISVs are going to say at the DR, before the 
DR, so that you don't get any rude surprises It is best (although not 
always possible) for the evangelist to meet personally with each DR 
attendee before the DR – perhaps hand-delivering the spec to be 
discussed at the DR. At this pre-DR meeting, the evangelist can go over 
the spec at a fairly high-level, and get a feel for what the ISV is 
likely to say or do at the DR. If the ISV's comments are likely to be 
positive, then the ISV can be encouraged to speak up at the DR, and the 
speakers can be encouraged to call on that attendee. If the ISV's 
comments are likely to be negative, the evangelist can try to have them 
addressed before the DR. They could be addressed, for example, by 
revising the spec – or perhaps by lining up other ISVs to oppose the 
negative comments at the DR (so that Microsoft is seen to be bowing to 
the will of other ISVs, rather than ramming its spec down the negative 
ISV's throat).

       Also, you may want to work closely with one or two "showcase" 
ISVs, to get them to demo support for the technology at the DR itself. 
This is risky. On the one hand, it shows that the technology really 
works, can be implemented relatively quickly, and already has momentum – 
perhaps instilling fear in the other attendees of being "left behind" if 
they don't catch up quickly with the "showcases." On the other hand, it 
will irritate the ISVs that were not chosen to be "showcases;" they will 
feel slighted.

       Invitations to DR's should be personalized, and sent out via 
email three weeks in advance, plus or minus a week or so. (Microsoft 
Word has a great mail-merge to email feature.) Any less warning and ISVs 
will have prior commitments. Also, our expcting developers to drop 
everything to come to us in very short notice, reinforces the notion 
that Microsoft is arrogant and demanding, among exactly the group that 
we most want to have thinking of us as being helpful and responsive. It 
is very insulting to ask someone to come out on less than two weeks' notice.

       Invitations should go out to a friendly subset of Tier A ISVs and 
to providers of evangelical infrastructure (see above). Press should 
rarely be invited. Analysts should only be invited if they are already 
bought and paid for. PSS should send the staff that they have wisely 
assigned to cover this emerging technology.

       3: Host the Design Review

       The goals of a Design Review are described above. Design Reviews 
are relatively informal by nature – no one wants or expects Robert's 
Rules of Order to be followed. However, the presenters should be 
properly prepared, the demos should not crash (too often), the food 
should be good (and in the right quantity), etc. The Evangelist should 
host the Design Review. The responsibilities of the host include
           * Greeting ISVs as they arrive, and encouraging other 
evangelists to greet their accounts, too

       MS-PCA 1913139
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 5 of 27
           * Making sure that the guests' random problems (rental car 
trouble, flight change issues, ripped pant seats, sudden illnesses, 
etc,) are dealt with promptly and with discretion
           * Introducing each speaker, and making sure that each speaker 
sticks to the schedule
           * Assisting with Q&A
           * Schmoozing like mad during breaks – gathering feedback from 
as many attendees as possible, and pointing attendees with tough 
questions or great suggestions at the appropriate speakers
           * Making sure that the speakers stick around for the breaks, 
meals, and evening events (if any)
           * Gather written, confirmed quotes from the ISVs for use in a 
post-DR press release (if any)
           * Generally doing anything else necessary (within the bounds 
of reason and taste) to ensure that the attendees get the information 
they need, give us the feedback that we need, and have a great time.

       Design Reviews should be chock-full of juicy technical 
information, with a smattering of market opportunity analysis and 
business tips. While many hardcore technoids would rather read a spec 
than a novel, this is not universally true, so Design Reviews should 
provide some non-technical fun, too. If the DR runs for two days, have a 
booze-n-schmooze event at some fun location on the middle evening.

       Most of all, the attendees should be constantly treated with 
respect. Each attendee should leave the event feeling that Microsoft 
really listened to what that attendee had to say.

       Usually, an ISV will see that its competitors are also attending 
the DR. This is a good thing. When an ISV sees its competitors in the 
room, the ISV gets nervous that maybe its competitors are already ahead. 
Try to schmooze with ISVs within eyesight of their competitors. Every 
laugh and handshake you share with an ISV's competitor will strengthen 
the ISV's resolve to regain the lead.

       Let the technology team spend their time explaining how the 
technology works. Spend your time listening to the ISVs explain what 
they think of the technology, and why. Take good notes, write them up 
with conclusions and recommendations, and forward them to the technology 
team.

       Be sure that each attendee receives the message clearly and 
unambiguously: We need, want, and appreciate their support for this new 
technology.

       4: Limited Developer's Release

       After the design review – sometimes, at the design review itself 
– the technology will be sufficiently stable to release to a small group 
of developers, under NDA. The design review attendees should be the 
first to get this release, although others may also get it. The goals of 
this release are
           * To generate feedback to the technology team on bugs, 
feature requests, documentation issues, and architectural issues that 
need to be addressed before the beta can go out to a wider beta group
           * To help bring PSS personnel up to speed on supporting the 
technology
           * To give book authors, consultants, contract houses, sample 
code writers, etc, something to get started with, so that they will be 
prepared to produce the necessary infrastructure when the technology 
goes into wider, non-NDA beta.
           * To help Microsoft understand the strengths and weaknesses 
of our technclogy vs. that of any competitors, such that we can prepare 
technical, marketing, and business-development responses
           * To support the argument that Microsoft is "open," given 
that Microsoft's internal product groups are almost certainly getting 
drops of the technology by this time.

       Starting with the Limited Developer's Release, Evangelism should 
be sitting in on all PM and bug- prioritization meetings, prepared to 
present data that supports the adjustment of bug priorities and feature 
implementation order, based on ISV feedback.

       Similarly, starting in this phase, Evangelism needs to be 
constantly monitoring the status of books, courseware, sample code, 
seminars, conference presentations, stacked panel discussions, and other 
elements of the evangelical infrastructure, to ensure that they are 
progressing towards availability at the time of general, non-NDA beta.

       MS-PCA 1913140
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 6 of 27

       The Limited Developers Release needs to be made available on a 
secure web server to which only participating ISVs have access, The 
server needs to be updated with new drops of the technology whenever 
they become sufficiently stable. Other means of electronic communication 
- private newsgroups/forums, list servers, bug & feature submission 
mechanisms, etc. – should be set up for use at this time, with an eye 
towards their scaling up as the number of beta sites increases over time.

       The Early Adopters (or FirstWave) program for the technology is 
finalized during this period (initial carrots and sticks having been 
proposed in the very first evangelism phase). A FirstWave program offers 
specific Tier A ISVs the opportunity to get specific technical and 
marketing benefits in return for their commitment to delivering support 
for the technology in applications by a time. All elements of the 
program are laid out very specifically in a written document- a 
Memorandum of Understanding, also known as a Letter of Agreement or LOA. 
The LOA must be signed by an Officer ofthe ISV - a Director, VP, CEO, or 
President - to ensure that the commitment is taken seriously by the ISV. 
If a PM signs it, the ISV can always claim later that the PM didn't have 
the authority to do so. Similarly, the LOA needs to be signed by 
Microsoft's Director of Evangelism. LOA's are not legal, binding 
contracts, and thus give each party wiggle room. They need to be 
positioned as clarifying documents, that ensure that there are no 
misunderstandingns as to what was being agreed upon, and by whom. We 
can't enforce an LOA, but we can certainly remember who has honored 
them, and who has not.

       An alternative to an LOA is a legally binding contract. For most 
ISVs – with whom Microsoft has a long- standing relationship – a legally 
binding contract is overkill. For some ISVs – with whom our relationship 
may have been equally long, but somewhat more tempestuous – a legal 
contract may be required. There are other reasons why you might want to 
have a legal contract, but there's one big reason why you don't want to 
do so: because then you're going to have to deal not only with Microsoft 
Legal, but with every other company's legal departments, too. This will 
not be fun, and it will consume vast quantities of time that could be 
better spent on evangelism rather than legal wrangling.

       A sample LOA can be found at the end of this document[2]

       It is imperative that Evangelism work closely with the marketing 
side of DRG in hammering out what marketing incentives can and cannot be 
promised to ISVs in the LOA. It's their budget, their marketing 
resources and their time that we're promising. The Marketing folks have 
a lot of experience in this area, and can be counted on to help come up 
with ideas and implementations that will work for everyone.

       5: Jihad

       A Jihad is a road trip. in which an evangelist visits a large 
number of ISVs one-on-one to convince them to take some specific action. 
The classic Jihad is one focused on getting Tier A ISVs to commit to 
supporting a given technology by signing the technology's Letter of 
Agreement (LOA - see above).

       A Jihad focuses on the Travelling Salesman aspect of evangelism. 
As in sales, the purpose of the exercise is to close – to get the mark 
the ISV to sign on the dotted line, in pen, irrevocably. Not to get back 
to us later, not to talk to the wife about it, not to enter a three-day 
cooling-off period, but to get the ISV to sign, sign, sign.

       If the start of the meeting is the first time the ISV has seen 
the LOA, then he's not going to sign it at the end of the meeting. Since 
we're asking for a very serious commitment we want the ISV to give their 
signing serious consideration. If the ISV cannot deliver, then his 
committing to deliver is worse than useless – the ISV's participation 
may occupy one of a limited number of available slots, keeping some 
other ISV from participating.

       To maximize the chance of getting the ISV to sign during the 
Jihad visit, make sure that

       [2] Thanks to Will Gregg.

       MS-FCA 1913191
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 7 of 27
           * The ISV has seen the LOA at least a week before the Jihad visit
           * The LOA is very clear about what exactly each side is 
promising to deliver, and when
           * An Officer of the ISV's corporation will be attending the 
meeting
           * Microsoft's Director of DRG has positioned the LOA with 
sufficient seriousness, in a cover letter or other communication in 
advance of the meeting
           * You make it clear from the start that the purpose of your 
visit is to answer any questions that they might have, preparatory to 
signing the LOA while you're there
           * They understand that those who do not sign the LOA, are 
frozen out of all further information about the techology until it goes 
into public beta
           * They understand (without being crude about it) that you 
will be making the same offer to their competitors
           * You have T-shirts or other swag to give to those who sign. 
lt's amazing what some people will do for a T-shirt.

       There are a million tips and tricks to effective road trips, and 
to being a Road Warrior in general, all of which is beyond the scope of 
this discussion.

       6: FirstWave Program

       For the duration of the FirstWave Program, the key evangelism 
task is bringing the participants' applications' support of the 
technology to a demonstrable state. This usually requires
           * A series of gatherings of the ISVs' key developers in the 
DevLab, at which the ISVs work on their code with the assistance of PSS 
and the technology dev team
           * Support from PSS, which helps them get trained for the 
higher volume of calls that they'll get when the technology enters 
public beta
           * Tracking the progress of each ISV's demo application(s) – 
ideally, on an internal web page - and consolidating the results into 
regular reports for management (also ideally on an internal web page)
           * Orchestrating regular drops of the technology to the ISVs 
(via a secure web site)
           * Planning and starting to execute on the co-marketing 
benefits promised in the LOA
           * Ensuring that bug fix and feature requests from ISVs are 
prioritized appropriately by the technology team
           * Driving and monitoring the development of the evangelical 
infrastructure (hooks, courseware, conference sessions, consultants, 
whitepapers, stacked panels, etc.)
           * Laying the groundwork for the technology to be supported by 
MSDN's one-to-many programs
           * Broadening exposure to the technology, within larger ISVs, 
beyond the targeted demo app to other product teams (be aware of the 
ISV's internal politics here!)
           * Planning and driving the event at which the ISVs' demos 
will be presented to key industry influencers (press, analysts, ISVs, 
allies, and competitors).

       The launch event should be timed to coincide with the 
availability of the technology's wide public beta, described below. This 
timing magnifies the effect of the event: events generate press. 
Developers read the press and learn that they can download the 
technology from a given URL – a specific action that they can take right 
now, which will help them start implementing the technology in their own 
applications right away, while it's still fresh in their minds (from the 
press coverage).

       7: Public Beta Release

       This is the first beta release for which an NDA is not required. 
Through the Web and MSDN, such releases can reach millions of developers 
– far more than Evangelism can possibly work with. At this point, the 
one-to-many programs must kick in, or Evangelism will be completely 
swamped and randomized. The wide beta release is a newsworthy event, and 
it must be milk[SIC] for all it's worth, because you won't get another 
such news opportunity until the final version ships. Gather the key 
industry influencers together at an event – either a stand-alone event, 
or an event that leverages some larger industry event – and hit'em

       MS-PCA 1913192
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 8 of 27

       with both barrels, grenades, atomic weapons, and the kitchen 
sink. Hold nothing back - do everything you can to overwhelm them with 
how complete, compelling, and inevitable the technology is. Show 
end-user benefits, developer benefits, the business case, the steak, the 
sizzle, and everything else - all in a short, punchy, psychologically 
devastating event. Ideally, everyone attending that event should walk 
away convinced that the game's over - Microsoft has already won. Have PR 
work with the press and analysts at and immediately after the launch 
event, to ensure that the event's coverage is as positive as possible.

       Upon the release of the public beta, Evangelism's job changes. It 
still needs to shepherd the Tier A apps from demo to delivery, but 
Evangelism also needs to shift gears, to guerilla marketing. One-to-many 
technology marketing is the job of MSDN, not DRG/Evangelism; let MSDN do 
it's job. Most ofthe slides, articles, whitepapers, and other materials 
developed during the preceding NDA phases, need to be handed over to the 
appropriate person or group in MSDN, so that they can be widely 
disseminated, clarified, and hammered repeatedly into the collective 
consciousness of developers everywbere.

       This next phase of evangelism – one of guerilla marketing - I 
term "the Slog."

       8: The Slog

       Guerilla marketing is often a long, hard slog.

           slog (s\+g) v. slogged, slogqing, slogs. –tr, To strike with 
heavy blows, as in boxing. -intr. 1. To walk with a slow, plodding gait. 
2. To work diligently for long hours. –n. . 1. long, hard work. 2. A 
long, exhausting march or hike. [Orig. unknown.] -slog'ger —American 
Heritage Dictionary, 1991

       In the Slog, Microsoft dukes it out with the competition. MSDN 
and Platform marketing are the regular forces, exchanging blows with the 
enemy mano a mano. Evangelism should avoid formal, frontal assaults, 
instead focusing its efforts of hit-and-run tactics

       In the Slog, the enemy will counter-attack, trying to subvert 
your Tier A ISVs to their side, just as you should try to subvert their 
ISVs to your side. New ISVs should be sought, and directed to MSDN's 
one-to- many programs. Evangelism should constantly be on the lookout 
for killer demos, hot young startups, major ISVs, customer testimonials, 
enemy-alliance-busting defections and other opportunities to demonstrate 
momentum for our technology. If bugs are found in our technology, or 
missing features are found to be critically important, then now is the 
time to identify and fix them. Stay engaged with the technology 
development team; ensure that you are a valuable resource for them, not 
a hectoring pest. Document all of your progress (ideally in regularly 
updated internal Web pages) and forward it regularly to management. If 
management is not aware of your progress, your successes, and your 
stumbling blocks, then they can't help. (They may not help anyway, but 
they can't if they don't know what you need.)

       Keep those Tier A ISVs on track to delivery! They are your 
strongest weapons and cannot be forgotten.

       The elements of the evangelical infrastructure - conference 
presentations, courses, seminars, books, magazine articles, whitepapers, 
etc. – should start hitting the street at the start ofthe Slog. They 
should be so numerous as to push all other books off the shelf, courses 
out of catalogs, and presentations off the stage.

       Working behind the scenes to orchestrate "independent" praise of 
our technology, and damnation of the enemy's, is a key evangelism 
function during the Slog. "Independent" analyst's report should be 
issued, praising your technology and damning the competitors (or 
ignoring them). "Independent" consultants should write columns and 
articles, give conference presentations and moderate stacked panels, all 
on our behalf (and setting them up as experts in the new technology, 
available for just $200/hour). "Independent" academic sources should be 
cultivated and quoted (and research money granted). "Independent" 
courseware providers should start profiting from their early involvement 
in our technology. Every possible source of leverage should be sought 
and turned to our advantage.

       MS-PCA 1913193
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 9 of 27

       I have mentioned before the "stacked panel". Panel discussions 
naturally favor alliances of relatively weak partners - our usual 
opposition. For example, an "unbiased" panel on OLE vs. OpenDoc would 
contain representatives of the backers of OLE (Microsoft) and the 
backers of OpenDoc (Apple, IBM, Novell, WordPerfect, OMG, etc.). Thus we 
find ourselves outnumbered in almost every "naturally occurring" panel 
debate.

       A stacked panel, on the other hand, is like a stacked deck: it is 
packed with people who, on the face of things, should be neutral, but 
who are in fact strong supporters of our technology. The key to stacking 
a panel is being able to choose the moderator. Most conference 
organizers allow the moderator to select the panel, so if you can pick 
the moderator, you win. Since you can't expect representatives of our 
competitors to speak on your behalf, you have to get the moderator to 
agree to having only "independent ISVs" on the panel. No one from 
Microsoft or any other formal backer of the competing technologies would 
be allowed – just ISVs who have to use this stuff in the "real world." 
Sounds marvelously independent doesn't it? In fact, it allows us to 
stack the panel with ISVs that back our cause. Thus, the "independent" 
panel ends up telling the audience that our technology beats the others 
hands down. Get the press to cover this panel, and you've got a major 
win on your hands.

       Finding a moderator is key to setting up a stacked panel. The 
best sources of pliable moderators are:
           * Analysts: Analysts sell out - that's their business model. 
But they are very concerned that they never look like they are selling 
out, so that makes them very prickly to work with.
           * Consultants: These guys are your best bets as moderators. 
Get a well-known consultant on your side early, but don't let him 
publish anything blatantly pro-Microsoft. Then, get him to propose 
himself to the conference organizers as a moderator, whenever a panel 
opportunity comes up. Since he's well- known, but apparently 
independent, he'll be accepted – one less thing for the 
constantly-overworked conference organizer to worry about, right?

       Gathering intelligence on enemy activities is critical to the 
success of the Slog. We need to know who their allies are and what 
differences exist between them and their allies (there are always 
sources of tension between allies), so that we can find ways to split'em 
apart. Reading the trade press, lurking on newsgroups, attending 
conferences, and (above all) talking to ISVs is essential to gathering 
this intelligence.

       This is a very tough phase of evangelism. You'll be pulled in 
every direction at once, randomized by short- term opportunities and 
action items, nagged by your Tier A ISVs and pestered by every other ISV 
that wants to become a Tier A. Management will want to know right now 
how you're going to respond to some bogus announcement by some random 
ISV. Some PM over in Consumer will demand that you drop everything to go 
talk to an ISV in Outer Mongolia, that's run by an old college chum of 
his. Competitors will make surprise announcements, lie through their 
teeth, and generally try to screw you just as hard as you are trying to 
screw them.

       Of course, if you are very, very lucky, there will be no 
competition to your technology. But this is almost never the case. ODBC 
had its IDAPI, OLE had its Openboc, COM had its SOM, DCOM has its CORBA, 
MAPI had its VIM, etc., etc., etc. The existence of a Microsoft 
technology nearly guarantees that a competitive technology will spring 
into existence overnight, backed by an impromptu association of 
Microsoft competitors which have decided to draw yet another Line in the 
Sand ("If we don't stop Microsoft here, then they are going to take over 
the whole world!").

       Without a competing technology to fight, you just hand everything 
over to MSDN, give your Tier A ISVs to PSS, and find a new technology to 
evangelize. But that takes most of the fun out of the game :-)

       9: Final Release:

       Evangelism of a given technology usually ends with the final, 
shipping release of that technology. One last big press event, with 
demos, a tradeshow, press releases, etc., is often called for, 
showcasing the apps that are sim-shipping and the customers that are 
using them. In the face of strong competition, Evangelism's

       MS-PCA 1913194
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 10 of 27

       focus may shift immediately to the next version of the same 
technology, however. Indeed, Phase 1 (Evangelism Starts) for version x+1 
may start as soon as this Final Release of version X.

       1O: Critical Mass

       The Slog may continue beyond the Final Release, for many months, 
until Critical Mass is reached. It is possible that Critical Mass will 
not be reached at all for Version X of a technology, such that Phases 
1-9 will have to be repeated – possibly more than once – before ever 
reaching Critical Mass.

       Critical Mass is reached when the technology starts evangelizing 
itself. When reviews subtract points if it's not supported; when 
analysts say "great product plan, but what about [Technology Name]?"; 
when VC's won't fund a company unless it supports [Technology Name] - 
that's Critical Mass. At that point, Evangelism of the technology stops, 
and Evangelism's resources are applied to other technologies – or, if 
you're lucky, moves into the Mopping Up phase.

       11: Mopping Up

       Mopping Up can be a lot of fun. In the Mopping Up phase, 
Evangelism's goal is to put the final nail into the competing 
technology's coffin, and bury it in the burning depths of the earth. 
Ideally, use of the competing technology becomes associated with mental 
deficiency, as in, "he believes in Santa Claus, the Easter Bunny, and 
OS/2." Just keep rubbing it in, via the press, analysts, newsgroups, 
whatever. Make the complete failure of the competition's technology part 
of the mythology of the computer industry. We want to place selection 
pressure on those companies and individuals that show a genetic weakness 
for competitors' technologies, to make the industry increasingly 
resistant to such unhealthy strains, over time.

       12: Victory

       Some technologies continue as competitors long after they are 
true threats - look at OS/2, the Operating System that Refused to Die. 
It is always possible - however unlikely – that competitors like 
OpenDoc, SOM, OS/2, etc, could rise from the dead... so long as there is 
still development work being done on them. Therefore, final victory is 
reached only when the competing technology's development team is 
disbanded, its offices reassigned, its marketing people promoted, etc. 
You have truly and finally won, when they come to interview for work at 
Microsoft.

       Victory is sweet. Savor it. Then, find a new technology to 
evangelize &emdash; and get back to work :-)

       MS-PCA 1913195
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 11 of 27

       Strategic Plan

       Apple Computer, WLM, Windows NT and the Mac OS

       James Plamondon
       September 21, 1995

       Objective

       The objective of this strategy is to convert Apple Computer, 
Inc., from OS foe to hardware friend.

       Executive Summary

       This strategy involves two coordinated efforts:

          1. WLM: We make WLM the most compelling solution for the 
deployment of Mac OS- based applications. If successful. this effort 
will establish Win32 as the de facto API of the Mac OS.
          2. Windows NT/PowerPC: We make Windows NT the most compelling 
operating system for PowerPC-based servers and workstations – 
specifically, for high-end graphics and publishing applications. If 
successful, this effort will allow Windows NT to usurp the Mac OS on 
Apple's own high-end hardware.

       Pros and Cons

       Positive result: Most cross-platform applications will be written 
first for Windows, and work best on Windows.

       Negative result: Many apps that might otherwise have been 
Windows-only, will be ported to the Mac using WLM – but this strengthens 
Win32 on the Mac, and thus the positive result.

       MS-PCA 1913196
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 12 of 27

       WLM

       With its new support for the PowerMac and OLE, WLM is poised to 
become a major success on the Macintosh. An independent consultant who 
previously recommended VC/Mac 2,0 to only 20% ofhis clients, now 
recommends VC/Mac 4.0 to 90% of them – and has told Apple that he 
expects 50% of the units sold on the Mac in 12 months to be using WLM. 
Apple itself is paying a consultant to port PeopleSoft to the Mac using WLM.

       Unable to offer developers a cross-platform development 
alternative to WLM, Apple wants to make sure that WLM-based applications 
take the best possible advantage of unique Mac OS features (such as 
Apple Guide, PowerTalk, QuickDraw GX, etc.). Apple hopes that this will 
prevent the Mac from becoming awash with WLM—based apps that treat the 
Mac as a least- common-denominator platform, which would just speed the 
demise of the Mac OS.

       We don't care whether Apple succeeeds in its attempt or not. 
Their endorsement of the Win32 API on the Mac effectively ensures that 
ISVs write their apps first for windows; that the resulting apps work 
best on Windows; and that the MacOS is the recipient of second-class, 
ported apps.

       While the availability of WLM will cause many apps that might 
otherwise have stayed Windows-only to support the Mac too, each of these 
ported apps strengthens the WLM standard – which in turn increases the 
attractiveness of writing to Win32, and hence writing first and best for 
Windows.

       Our efforts with WLM are only one flank of a double-envelopement 
maneuver – the other flank being Windows NT on the PowerPC (see below). 
Windows NT/PowerPC comes at the Mac OS from the high end down; WLM comes 
at the Mac OS from the low end up. Therefore, VC/Mac and WLM should not 
focus solely on the Mac's high end; they must continue to target 
lower-end systems, to, Specifically, VC/Mac should continue to support 
the 68K Mac.

       Tactics:
           * Apple: Get Apple to publicly endorse WLM; to use it in 
their own cross-platform development; to assist in its development; to 
evangelize its use hy ISVs; to make Copland like WLM; to include WLM in 
Copland.
           * DAD/CSD: Use WLM in our own apps, especially those that are 
likely to be bundled with Macs, given away free, or otherwise have high 
penetration; ship as few versions of WLM as possible; ship the WLM DLL 
file intact and unaltered; drop the licensing restrictions on WLM.
           * PSD/BSD: Help keep WLM in sync with additions to Win32; 
work with Apple to make NT/SFM an even better player in Mac networks; do 
what is necessary(technically and marketing) to take the graphics and 
publishing business away from the Mac.
           * DD: Work closely with Apple and Microsoft's internal groups 
to support WLM; license WLM directly to Mac tool vendors (after or 
instead of bundling); keep up with additions to Win32; continue support 
for 68K.

       MS-PCA 1913197
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 13 of 27

       Windows NT on the PowerPC

       As Apple's PowerMac converge with IBM and Motorola's PReP 
machines onto the CHRP standard, we have a golden opportunity to set 
Apple's strategy on its head, and usurp the Mac OS on Apple's own 
hardware.[3]

       CHRP boxes from the current PReP vendors will run Windows NT very 
well; NT is their focus. So Apple will be faced with direct competitors 
that offer equivalent hardware, running Windows NT instead of the Mac OS 
- and Windows NT apps instead of Mac OS apps. If these Windows NT/CHRP 
applications have a price/performance advantage over the equivalent Mac 
OS/CHRP apps, Apple's going to be in a very tough competitive position. 
And given that the Mac OS won't support threads (let alone symmetric 
multi-processing) for over a year, it seems likely that-all else being 
equal—Apple is indeed likely to be in a tough spot, at the high end, at 
least. SUR's lack of Plug-n-Play, and steep hardware requirements, will 
keep it off of the desktop in the home and education markets,leaving 
those to the Mac OS for a while longer.

       At the high end, though – workstation and server apps – 
NT/PowerPC market is about to explode, due to the confluence of three 
factors:
          1. SUR: The Shell Update Release is going to make Windows NT 
more palatable to everyone – especially those who might otherwise choose 
a Macintosh The wave of PR/Marketing surrounding its release, while 
being nothing like the Windows 95 hype (I assume), will do much to 
increase the acceptance of NT as a desktop system. And the more rapidly 
Windows NT takes off, on any chip, the better for Windows NT on other chips.
          2. VC/NT/PowerPC: The release of Visual C++ 4.0 for Windows 
NT/PowerPC, at year's end, will make it easy for the developers of 
Windows 95 applications to meet the demand for the PowerPC-native 
versions that SUR will generate.
          3. CHRP: Apple adopts a hardware standard that supports 
Windows NT, at just exactly the same moment that Windows NT's user 
interface is improved beyond that of the Mac OS. (You gotta love it.) 
Apple (and its licensees), IBM, Motorola, and the Taiwanese, 
collectively ships millions of Windows NT-capable boxes.
       To make it possible for a mixed shop to standardize on Windows 
across the board, we'll need a version of Windows NT that can run on 
Apple's installed base of PowerSurge Macs. Motorola says that this work 
is not theoretically difficult, although they are concentrating their 
attentions on CHRP first, which is fine. We also want to lower the cost 
of switching from the Mac OS to Windows NT/PowerPC, given hardware that 
can run both 0Ss; this can be done (in part) be encouraging ISVs 
(including Microsoft's apps groups) to include the NT/PowerPC version of 
an app on the PowerMac SKU of that same app (as well as on Windows SKU). 
Tactics:
           * Apple: Work with Apple to ship NTS on Shiner; to improve 
the support for the Mac OS in NT/SFM.
           * Mac OS Licensees: Work with the licensees to ensure that 
their CHRP and Powersurge machines run Windows NT really well (dual-hoot).

       MS-PCA 1913198
       HIGHLY CONFIDENTIAL

       [3] Doing the same to IBM's OS/2 on PowerPC strategy at the same 
time – slaying two birds with one stone.
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 14 of 27

           * DAD/CSD: Produce complete, optimized NT/PowerPC versions of 
every app for which you currently plan to ship either a Mac OS or an 
NT/Intel version, including support for DAO, Jet, etc.; ship the 
NT/PowerPC version on both the Windows CD SKU md thc PowerMac CD SKU. 
(Except for really low-end home/education Mac OS stuff)
           * PSD/BSD: Work with Apple to make NT/SFM an even better 
player in Mac networks; do what is necessary (technica11y and marketing) 
to take the graphics and publishing business away from the Mac; produce 
Intel-peer, complete and opdmized versions of NT, BackOffice, etc.
           * DD: Produce Intel-peer, complete and optimized versions of 
our entire tools suite, including DAO, Jet, etc., keeping in sync with 
all Intel releases, etc.

       MS-PCA 1913199
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 15 of 27

       Appendix A: Windows Layer for the Macintosh (WLM)

       WLM is Microsoft's implementation of the Win32 API on top of the 
Mac OS. A developer can take a Win32-based application's code, recompile 
it, and link it to WLM, with the result being an application that looks 
and feels like a native Macintosh application – that was written 
entirely to the Win32 API.

       The wide adoption of WLM is good for Microsoft, because it 
ensures that applications are written first for Windows, and work best 
on Windows, with the Mac as a second-class platform.

       WLM is currently available only as an element of Microsoft Visual 
C++ Cross-Development Edition for the Macintosh, but it will also be 
available soon in a separate WLM/MPC SKU. The success of the WLM SKU is 
nearly assured, as its creation was driven by the desires of the Mac 
compiler vendors (Metrowerks and Symantec).

       WLM's strengths were obscured by its lack of support for OLE and, 
especially, the PowerMac, in its first release (in VC/Mac 2.0, November, 
1994). These features are being added to the forthcoming version, to be 
released in VC/Mac 4.0 (November, 1995) WLM 4.0 still lacks support for 
multimedia and OLE Controls, which are currently being investigated for 
support in next year's release

       MS-PCA 1913200
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 16 of 27

       Appendix B: The PowerPC, PReP, CHRB and Copland

       The PowerPC is a RISC-based CPU manufactured by IBM and Motorola [4]

       There are two different standards for personal computers based on 
the PowerPC CPU: the PowerPC Reference Platform (PReP), ad Apple's PowerMac.

       All od Apple's high-end Macs, and an increasing share of its mid- 
and low- end Macs, are based on the PowerPC. All PowerPC-based Macs are 
referred to as "PowerMacs." Early PowerMacs were hard-wired to be 
big-endian (as required by the Mac OS), and thus could not run Windows 
NT (which requires a little-endian architecture). Older PowerMacs also 
used the NuBus bus architecture. The newest PowerMacs, however, are 
endian-switchable at run- time, and use the PCI bus, so that they could 
conceivably run Windows NT in addition to the Mac OS. These newest 
PowerMacs -the 8500 and 7500, code-named the PowerSurge line - can be 
viewed as a half-step by Apple towards CHRP-compliance (see below).

       Apple has licensed its PowerMac hardware designs (including 
PowerSurge) and the Mac OS itself to three small companies (Power 
Computing, DayStar, and Radius), which are now shipping Mac compatibles. 
These Mac OS licensees are very interested in offering Windows NT on 
their future CHRP systems, both to hedge their bets against the 
potential failure of the Mac OS, and to differentiate their Mac 
compatibles from Apple-logoed Macs.

       IBM, Motorola, FirePower, and others, manfacture PowerPC 
computers that are based on thc PowerPC Reference Platform (PReP) 
standard, defined by IBM and Motorola. All ship with Windows NT 
pre-installed.

       The Common Hardware Reference Platform (CHRP) is the unification 
of the PowerMac and PReP designs, into a single hardware standard that 
is intended to run Windows NT, the Mac OS, and others (OS/2, Solaris, 
AIX, NetWare, NextStep, etc.). CHRP is based on standard industry 
components - for everything except the core PowerPC CPU. CHRP's backers 
hope to make it a strong second standard, challenging the Intel 
chip/chipset/motherboard monopoly CHRP is also backed strongly by the 
Taiwanese New PC Consortium (TNPC), whose board—manufacturing businesses 
have recently been idled by Intel's entry into that market.

       The CHRP spec is expected to become final late in 1995. Motorola 
has Windows NT up and running on CHRP prototype today, which they plan 
to demonstrate at Fall Comdex in November 1995. CHRP systems running 
Windows NT ar expected to hit the market as soon as as Q1 96.

       Apple has stated that it will ship its first CHPP-compliant 
computers in 1ate l996, but much of tbeir tardiness is due the 
expectation that these machines will be running the MacOS, which is not 
expected to run on CHRP-compliant systems until that time. Apple may be 
able to ship a CHRP box that runs Windows NT well before it can ship one 
runn ing the MacOS (just as IBM is now shipping NT on its PReP 
computers, while OS/2 for PowerPC languishes in development),

       "Copland" is the codename for the next version ofthe Mac OS, 
a.k.a. "System 8", designed specifically to take advantage of the CHRP 
standard. Apple is saying publicly that Copland will be ready in late 
1996, in time for their first CHRP systems – but it is widely believed 
that Copland will not be released until mid-to-late 1997. It is known 
that Apple is porting their existing Mac OS -System 7.5 -to CHRP, in 
case Copland does not ship on schedule.

       [4] Versions 60l, 603, and 604 are shipping now; the 610 is under 
development; the project to build the 615 - rumored to have 486 
compatibility in silicon - has apparently been scrapped.

       MS-PCA 1913201
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 1 of 27

       Appendix C: Apple and Windows NT

       The Mac OS is a crummy OS for servers, and Apple knows it. On the 
other hand, Windows NT is doing very well on servers, and its support 
for Mac file and print sharing[5] makes it a natural server for 
Mac-centric environments.

       At one point in late 1994, Apple's Director fof CHRP Marketing, 
Jim Gable, was running n skunk-works project with Motorola, with the 
intention of implementing Windows NT on its PowerSurge-based "Shiner" 
server. When Spindler heard about this skunkworks effort early this 
year, he went ballistic, and shut it down.

       Sinee then, Apple has reorged, and Don Strickland — formerly of 
Mac OS Licensing — has taken over as VP of Apple's enterprise marketing. 
He is backing a new effort to get NT as the default OS on "Shiner" 
(which has been redesigned to be fully CHRP-compliant), and has secured 
the approval of Apple`s President and CEO, Michael Spindler. Microsoft, 
Motorola, and Apple met to discuss this new effort the week of September 
25th, and agreed that it could be ready to ship in Q1 96. The shipment 
of an Apple-logoed server with windows NT on it will be opposed by 
Apple‘s Mac OS group, and they may succeed in killing tbe project-but 
Don Strickland's backing will probably help Shiner ship with NT on board.

       [5] Windows NT Server has a collection of “Services for 
Macintosh" that make it, as Stewart Alsop said in InfoWorld., "The 
obvious server for Macintosh networks."

       MS-PA 1913202
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 18 of 27

       Appendix D: Apple's Down - But Not Out

       Declining ISV support. Almost 90% of Mac ISVs are also writing 
for Windows. Apple is having pay Windows ISVs such as PeopleSoft to 
produce Mac versions of their applications - and even then, PeopleSoft 
is using WLM, rather than writing native Mac OS code. Metrowerks, the 
leading Mac tool vendor, recently released a Mac-hosted, 
Windows-targeting cross-compiler – to a standing ovation – that will 
accelerate this migration to Win32, and they are in negotiations to 
license WLM as well. Adobe recently blamed its poor financial results on 
its over-dependence on the Mac, and is moving dramatically towards 
Windows and Windows NT. 0penDoc – a key element of Apple’s system 
software strategy – is rapidly losing mindshare; its failure will 
seriously discredit the Mac OS group within Apple.

       Declining Market Share. The Mac OS’s share has dropped to having 
under 8% of the OS market, and analysts project it to continue to spiral 
downwards. SPA recently reported that sales of Mac applications were 
down 7% in the second quarter of '95, compared to '94. The increasing 
sales of Windows 95 will push the Mac OS' market share even lower by 
early next year.

       Declining Profitability. Meanwhile, Apple has announced a new 
line of its PowerPC-based PowerMacs – the PowerSurge line – that offer 
all the power of their previous configurations, at much lower cost. 
Sales of their older high-end Macs have slowed, awaiting the new 
PowerSurge Macs – which are unavailable, due to part shortages. As a 
result, Apple recently announced a dramatic 48% decline in profitability.[6]

       Declining Technical Advantage. Because Windows 95 has reached 
technical parity with the Mac OS (roughly speaking), Apple can no longer 
charge a premium for the Mac – it has to be price-competitive with IBM, 
Compaq, and the other first-tier Wintel vendors. But unlike its PC 
competitors, Apple has a huge software R&D budget, devoted to 
maintaining and advancing its proprietary OS. The next version of the 
Mac OS is expected to be released in 1997 – 18 months or more away. Even 
then, the new version will not offer features drmnatically superior to 
Windows 95, let alone Windows NT, And Windows will, by then, have 
extended its lead even further – in quality, in installed base, and in 
application support. Finally, a forthcoming report from IDC shows that 
users are more productive using the Windows 95 UI than the MacUI – 
robbing the Mac OS of its one last claim to superiority.

       Entrenched Base. Apple does have an installed base of around 16 
million active users, mostly in the home, education, and in graphics and 
publishing. The home and education units are going to remain in place 
well into the next millenium; those markets just don‘t turn over their 
hardware quickly – so there will be a continuing (albeit declining) 
demand for Mac0S eoftware for years to come. On the other hand, the 
graphics and publishing market turns over its hardware very quickly, 
since they are usually riding the very leading edge of the power curve – 
so Windows NT Workstation should be able to blow the Mac OS off of 
Apple's own hardware in this market in pretty short order.

       Apple's Spring Offensive. Despite its many reverses, Apple is not 
yet dead – nor is the Mac OS. Apple has reorganized its developer 
support efforts, and can now evangelize and support its few remaining 
developers efficiently. When a developer release of Copland becomes 
available early next year, Apple will have something tangible to 
evangelize, giving the Mac OS lots of new mindshare. And OpenDoc will 
finally be released in final form, to great fanfare. The impending 
availability of CHRP-compliant Macs will also encourage a new crop o£ 
Mac Os licensees to sprout at about this time.

       [6] The "Osborne Effect" in action. When Adam Osborne 
enthusiastically described the features of his forthcoming Osborne II 
computer, sales of the existing Osborne I fell so sharply that his 
company went bankrupt before the new model could be brought to market.

       MS-PCA 1913203
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 19 of 27

       In short, we should expect to see a major Spring Offensive from 
Apple in 1996, which we will need to blunt and turn back.

       MS-PCA 1913204
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 20 of 27

       LETTER OF AGREEMENT (LOA)
       BETWEEN MICROSOFT CORPORATION ("MS")
       AND ___________________________ ("COMPANY")
       DATE _____________________

       Microsoft® Developer Relations Group Internet Program for 
Internet Authoring Tools

       This LOA sets out the agreement between MS and COMPANY with 
respect to the development of authoring tools using MS Internet client 
and server platform technology as described helow. The LOA, in 
conjunction with attached material, describes the MS Developer Relations 
Group Internet program for Internet authoring tools, its benefits, and 
requirements for participation through September 1997, or the release or 
Internet Explorer 4.0, whichever is later.

       In brief, this program provides COMPANY with technical asistance 
in implementing exciting new features in COMPANY's Internet authoring 
tools, as well as co-marketing assistance in promoting its Internet 
authoring tools. The benefit to MS is in having Internet authoring tools 
for creating compelling, exciting products, which will showcase new 
features of MS's Internet solutions.

       Program Benefits - Technical Assistance from MS
          1. MS will provide premier-level product support to COMPANY 
for the Internet technologies named in this document, freee of charge.
          2. MS will provide COMPANY access to the latest releases 
(including interim builds) of Internet related platforms, products and 
development tools, in accordance with their respective beta testing 
agreements, through a private web site prior to any broad software release.
          3. MS will provide a special "technical envoy" from MS's 
Developer Relations Group to assist in clearing roadblocks, which might 
arise during the development of COMPANY's Internet authoring tools, 
including scheduling meetings between the COMPANY and Microsoft at 
Microsoft's discretion and subject to time constraints, schedules and 
availability.
          4. MS will invite COMPANY for special implementation lab weeks 
and workshops in Redmond, providing COMPANY great support for developing 
and debugging its Internet authoring tools. Travel and lodging expenses 
are COMPANY's responsibility.

       Program Benefits — Co-marketing Assistance from MS
          1. MS will provide infomation on COMPANY's Internet authoring 
tool and their support for the features described in "Requirements for 
Participation" (below), or demonstrate COMPANY's Internet authoring 
tools to leading Independent Content Providers (ICPs) participating in 
the Developer Relations Group Internet Program.
          2. MS will invite COMPANY to exhibit in a Microsoft-affiliated 
area at targeted conferences, in accordance with the terms of the 
applicable Exhibitor Agreement. Possible conferences include Comdex 
Canada (Toronto, Ontario, July 9-11), Internet World/Summer (Chicago, 
IL, July 21-Z5) and Microsoft's Professional Developers Conference (San 
Diego, CD, September 23-26). All conferences are tentative and subject 
change.
          3. MS will invite COMPANY to propose a demonstration of its 
internet authoring tools to be included in a live satellite broadcast 
("Microsoft at the Movies") focused on Internet Explorer 4.0. possibly 
in conjunction with the Microsoft Professional Developers Conference.

       MS-PCA 1913205
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 21 of 27

          1. MS will invite COMPANY to include a demonstration version 
of COMPANY's Internet application tools on a CDROM, if such a CD-ROM is 
produced. Over 12,000 attendees received CD-ROMs at the most recent 
oomparable event
          2. COMPANY's Internet authoring tool will have the opportunity 
to be featured on Microsoft's Site Builder Network. Terms and conditions 
of the agreement for featuring COMPANY's Internet authoring tool have 
not be finalized and are subject to Site Builder Network approval.
          3. COMPANY's Internet authoring tool will be featured for one 
or more weeks in the new "Tool of the Week" section of Microsoft's Tools 
Gallery. (currently located at 
http://www.microsoft.com/gallery/files/tools/) Within 60 days following 
the availability of COMPANY's pre-release or released Internet authoring 
tool complying with "Requirememts for Participation" (below).
          4. COMPANY will have the opportunity to license selected 
content from MS's Site Builder Network Galleries, for inclusion with 
COMPANY's internet authoring tools. Terms and conditions of this license 
have not been finalized.
          5. MS will make best efforts to support any press releases or 
other public activity that COMPANY carries out relating to the Internet 
and the use of MS technology.
          6. MS may provide other co—marketing avenues from time to 
time, such as site contests and joint advertising. By being on this 
program, COMPANY will be among the first to be invited to participate in 
these co-marketing initiatives.

       Requirements for Participation
       Product Requirements

          1. COMPANY will obtain the "Designed for Windows NT and 
Windows 95 Logo" for its Windows-based Internet authoring tools. If 
COMPANY's Internet authoring tool is end-user focused COMPANY will 
obtain the "Office97 Compatible Logo", additionally, if COMPANY's 
Internet authoring tool adds value to the Microsoft BackOffice Suite, 
COMPANY will obtain the "Designed for Microsoft BackOffice Logo". 
Information regarding Microsoft's Logo programs is available at 
http://www.microsoft.com/windows/thirdparty/ and 
http://www.microsoft.com/office/compatible/. Microsoft will consider 
paying the expences of obtaining any logos for the COMPANY's Internet 
authoring tools, on a case-by-case basis, depending on the level of 
technology support offered by COMPANY's Internet authoring tools.
          2. COMPANY's Internet authoring tools will check inserted 
ActiveX controls and Java applets for digital certificates using the 
WinVerifyTrust interface, there by supporting Microsoft's Authenticode 
technology. For ActiveX controls and Java applets without digital 
certificates, or whose certificates are invalid, (out of date, revoked) 
COMPANY's Internet authoring tools will provide appropriate warnings. 
(See 
http://www.microsoft.com/msdn/sdk/platfoms/doc/sdk/win32/func/src/f91_32.htm 
for information on the WinVerifyTrust interface.) Additionally COMPANY's 
Internet authoring tools will use the IClassFactory2 interfaces to 
enforce ActiveX control licensing. (see 
http;//www.microsoft.com/msdn/platforms/doc/activex/src/comext_8.htm for 
information on the IClassFactory2 interfaces.)
          3. COMPANY will enhance its Internet authoring tools to host 
Design Time Controls, as described in the Web Design-Time Control SDK, 
available as http://www.microsoft.com/intdev/sdk/dtctrl/dcsdk.exe. 
Microsoft is developing a selection of Design Time Controls, and is 
considering licensing those controls for redistribution with 3rd party 
authoring tools. Possible controls include, data binding controls, CDF 
creation controls,

       MS-PCA 1913206
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 22 of 27

       Visual CSS positioning controls, and Multimedia Effects controls. 
The specific controls and licensing terms and conditions have not been 
finalized.
          4. COMPANY will enhance its Internet authoring tools to 
support Cascading Style Sheets (CSS) as defined in the Internet Client 
SDK at http://www.microsoft.com/sitebuilder/features/inetsdk.asp and by 
the W3C at http://www.w3.org/pub/WWW/Style. This support should include 
mpport for all CSS elements and attributes, and the ability for authors 
to visually change any element attributes.
          5. COMPANY will enhance its Internet authoring tools to 
Support CSS positioning as defined in the Internet Client SDK at 
http://www.microsoft.com/sitebuilder/features/inetsdk.asp and by the W3C 
at http://www.w3.org/pub/WWW/TR/WD—positioning . This support should 
include the ability for authors to position any element with CSS defined 
precision.
          6. COMPANY will enhance its internet authoring tools to 
support Internet Explorer 4.0 multimedia effects, as defined in the 
Internet Client SDK at 
http;//www.microsoft.com/sitebuilder/features/inetsdk.asp. This 
functionality should allow authors to define interactive, media rich web 
pages, by providing an easy interface for using effects such as 
transitions, sequences, paths, audio, complex graphics, spites, and the 
other effects provided with Internet Explorer.
          7. COMPANY will enhance its Internet authoring tools to 
support the creation of Channel Definition Format (CDF) files as defined 
in the Internet Client SDK at 
http;//www.microsoft.com/sitebuilder/features/inetsdk.asp and by the WSC 
at http://www.w3.org/pub/WWW/Style/. This support should include the 
ability for authors to select any content, including cross-site content, 
and then automatically generate the CDF file, and required META tag 
information.
          8. COMPANY will enhance its Internet authoring tools to 
support Internet Explorer 4.0 data binding attributes, as defined in the 
Internet Client SDK at 
http;//www.microsoft.com/sitebuilder/features/inetsdk.asp . Internet 
Explorer 4.0 data binding allows authors to asyncronously push data to 
specified HTML pages. This functionality should allow authors to select 
elements to bind to available data sources. Allowing authors to define 
the format of returned data, ie. repeated table, or current row.
          9. COMPANY will enhance its Internet authoring tools to 
support the PICS rating system, as defined by the WSC at 
http://www.w3.org/pub/WWW/TR/REC-PICS-services-961031.html. This support 
should include the ability to visual set the PICS settings, and then 
generate and insert the required META tag information.
         10. COMPANY's Internet authoring tools will support Active 
Server Pages (ASP). This support must include ASP shorthand syntax 
"<%...%>" and value added support for the server components that ship 
with IIS 3.0 or later. (See http://www.microsoft.com/iis/ for 
information regarding Active Server Pages and IIS components.)
         11. COMPANY will enhance its Internet authoring tools to add 
Microsoft Transaction Server (MTS) support to all server side components 
that ship with its Internet authoring tools, including Java applets. 
(See http://www.microsoft.com/transaction/ for information regarding 
Microsoft Transaction Server.)
         12. COMPANY will enhance its internet authoring tools to add 
value-added support to Microsoft's SiteServer and other server side 
components. (See http://wwv.microsoft.com/siteserver/ for information 
regarding Microsoft's SiteServer.)

       MS-PCA 1913207
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 23 of 27
         13. COMPANY's Internet authoring tool will host the Microsoft 
Internet Control (the Microsoft Internet Explorer HTML: engine, also 
known as "WebBrowser") as its integrated browser, there-by providing 
authors with integrated previewing capability. (See 
http://www.microsoft.com/workshop/prog/sdk/docs/iexplore/webrowse.htm 
for information on integrating the WebBrowser control.) Additionally 
COMPANY will license Internet Explorer 4.0 or later for redistribution 
with their Internet authoring tools. (See 
http://www.microsoft.com/ie/ieak for information on redistributing 
Internet Explorer.
         14. COMPANY's Internet authoring tools will host the Microsoft 
Active scripting interfaces, allowing interactive authoring and 
debugging of scripts and Java applets.
         15. COMPANY will enhance some Java components and applets that 
are redistributed with its Internet authoring tools to utilize 
Microsofts Application Foundation Classes, including Microsofts AFC 
Enterprise classes for some server based Java components. Additionally 
COMPANY will license and redistribute the AFC classes with its Internet 
authoring tools.
         16. All ActiveX controls and Java components that are 
distributed with COMPANY's Internet authoring tools will be stored in 
CAB files and digitally signed using Microsoft's Authenticode(TM) code 
signing specification (See http://www.microsoft.com/workshop/prog/cab/ 
for information on CAB tiles and 
http://www.microsoft.com/workshop/prog/security/authcode/codesign.htm 
for information on signing CAB files.)
         17. COMPANY's Internet authoring tools will support creation of 
Web Collections, a W3C standard derived from Microsoft's SiteMap 
technology. Web Collections are used to represent the Table of Contents 
and Keyword Index files used by the MS HTML Help technology.
         18. COMPANY will register their Internet authoring tools with 
the Microsoft Tools Gallery, currently located at 
http://www,microsoft.com/gallery/files/tools/ .
         19. COMPANY will register their internet authoring tools with 
tho Microsoft J/Advantage program. Information regarding the J/Advantage 
program can be found at http://www.microsoft.com/java/.
         20. COMPANY agrees to display marketing materials (e.g. small 
signs, monitor toppers) noting product's support for Microsoft's 
technologies (eg. Internet Explorer, Microsoft Java VM, AFC) whenever 
PRODUCT is being demonstrated publicly (e.g. booths at trade shows, 
conferences, seminars, roadshows.)
         21. COMPANY will provide by June 15th, a quote from its most 
senior executive stating COMPANY's support for the Microsoft Dynamic 
HTM1, another supporting CDF, and a third supporting both the Dynamic 
HTML and CDF, and the right to use these quotes in internet-related 
press releases for 60 days thereafter.
         22. COMPANY's Internet authoring tool will support for 
inserting the authorized Microsoft Internet Explorer logo on a page 
through an easy to use interface, in accordance with the applicable logo 
agreements. Additional information can be found at 
http://www.microsoft.com/ie/logo/.

       Site Requirements

          1. If COMPANY maintains its own web site, COMPANY will use the 
Windows NT Internet Information Server 3,0 or later as the primary http 
server on its site and host at least one page showing support for Active 
Server Pages.

       MS-PCA 1913208
       HIGHLY CONFIDENTIAL

       Generalized Evangelism Timeline 	Microsoft Confidential	Page 24 of 27

          2. COMPANY will add Microsoft to the list of "Technology 
Alliances and Partnerships" as listed at http://www.allaire.com/partners/.
          3. COMPANY will display the authorized Microsoft Internet 
Explorer logo on its default home page, in accordance with the 
applicable logo agreements. Additional information can be found at 
http://www.microsoft.com/ie/logo/. Or COMPANY agrees not to display any 
competitive product logos on their home pages.
          4. If COMPANY provides their Internet authoring tools for 
downloading from their web site the setup program will be digitally 
signed in accordance with Microsofts Authenticode(TM) code signing 
specification. (See 
http://www.microsoft.com/workshop/prog/security/authcode/codesign.htm 
for information on signing executable files.) NOTE: Site Builder Network 
requires all downloadable files to be digitally signed.

       Event Requirements

          1. COMPANY will use Internet Explorer 4.0 or later as the 
default browser for all demonstrations and keynotes at Microsoft 
sponsored events, and will give default to Internet Explorer 4.0 or 
later but giving equal use to other competing products at all 
non-Microsoft sponsored events.

       Time Frames

       COMPANY agrees to meet the following schedules in order to enter 
and ensure continued participation in this program:
          1. COMPANY should have completed requirement 1 listed under 
the "Site Requirements" section within two weeks of entering into this 
agreement. COMPANY should have completed all requirements listed under 
the "Site Requirements" section within one month of entering into this 
agreement.
          2. COMPANY should make a "Beta" release of its Internet 
authoring tools available for download via its web site within four 
months of entering into this agreement, meeting all requirements listed 
under the "Product Requirements" section.
          3. COMPANY will ship their Internet authoring tools within 
three months of the shipment of Internet Explorer 4.0. Please note that 
participation in some marketing activities may require a shorter and 
less flexible schedule as well as cross-platform versions of COMPANY's 
products. Microsoft will use the following COMPANY-supplied information 
to track company's eligibility (all data will be treated confidentially 
as described in the "Confidentiality" section):
       Product Name(s)	Platform(s) Supported	Product Description(s) 
Expected "WWW Beta"	Expected Final Shipment Date
        	 	 	 	
        	 	 	 	

       COMPANY WEB SITE NAME: 
_______________________________________________________________
       COMPANY WEB SITE URL: 
_______________________________________________________________
       CONTRACTORS USED (if any): 
___________________________________________________________

       Confidentiality

       MS-PCA 1913209
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 25 of 27

       MS and COMPANY shall keep in confidence the existence of this 
LOA, any of the terms and conditions described herein, and any 
confidential or proprietary documents or information relating to our 
products or programs. Neither MS nor COMPANY shall disclose any such 
information to a third party except as may otherwise be permitted in the 
existing Non-Disclosure Agreement between COMPANY and MS. dated 
__________________.

       Accepted and agreed to:
       By signing below, you are agreeing:
           * to the terms and conditions of the Ms Developer Relations 
Group Internet program for Internet Authoring Tools outlined in this letter
           * that you are am officer of your company with authority to 
enter into this agreement with MS
       CORPORATION 	MICROSOFT®
       By: _____________________________ 	By:
       Name: _____________________________ 	Name:
       Title:_____________________________ 	Title:
       Date: _____________________________	Date:

       Attachments:
       MS STANDARD NDA

       MS-PCA 1913210
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline Microsoft Confidential Page 26 of 27
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 26 of 27

       MS-PCA 1913211
       HIGHLY CONFIDENTIAL
       Generalized Evangelism Timeline 	Microsoft Confidential	Page 27 of 27


Ian Lynch wrote:
> On Fri, 2007-11-02 at 05:51 -0400, Jody Goldberg wrote:
>> On Wed, Oct 31, 2007 at 11:00:14PM +0000, Ian Lynch wrote:
>>> To answer Marbux, I think that the Gnome issue is different from the
>>> profit making companies because the values of a not for profit
>>> organisation are not the same as for profit making company. Generally
>>> for profit companies exist for the benefit of their shareholder, not for
>>> profits exist for some noble cause. Its quite easy to see why Gnome's
>>> involvement with ECMA would be seen as different from Novell, Sun or
>>> Apple by the community.
>>>
>>> Again, think people, think emotions, think perception not technical
>>> details. This is about human and community relationships, not coding or
>>> legal analysis. 
>> As an implementer I disagree with those priorities. 
> 
> They aren't priorities :-) Its interesting that you again rationalise
> something that is irrational by nature. The inability to understand the
> importance of the emotional dimension in human dealings is a good part
> of the reason why the Windows and Office mess has persisted for so long.
> 
>> This is about technical details, not perception.
> 
> Just saying it isn't going to change the outcomes. The fact is that
> people are emotional even if you think they shouldn't be. Organisations
> within the community are made up of emotional beings and are
> inextricable linked to their emotions. Digital freedom has the emotional
> power to make people devote their lives to it. 
> 
>> This is about writing the code necessary to get something done.
> 
>> The gains from having better interoperability with MS (OOX and
>> binary formats) outweigh the miniscule political cost.
> 
> You under-estimate the political cost because you are seeing it from
> your own political perspective. This discussion illustrates the fact
> that many people in the FOSS community are likely to see it differently.
> They might all be being irrational but in the end its the outcome that
> matters.
> 
>>  As much as I
>> enjoy Gnumeric, the notion it is going to materially alter the
>> decisions by the national bodies in ISO the decision by ISO seems
>> ridiculous.
> 
> You seem to have missed the entire point. The political symbolism
> reaches far beyond Gnumeric. If it was just about Gnumeric and its take
> up I would agree with you, if Gnumeric disappeared tomorrow it would
> have no effect on my life as far as I can tell but if there is a very
> fine margin between getting ISO status and not for OOXML and MS are able
> to use any small amount of leverage from gaining FOSS credibility then
> there is a finite chance that its that that tips the balance. Then it
> affects everyone in the community, not just users of Gnumeric.
> 
>> On the other hand it seems certain that there are already people who
>> will benefit from being able to use Gnumeric in place XL thanks to
>> improvements in our filters.
> 
> And they couldn't do that for the next 6 months using XLS files exported
> from the relatively few people that use Office 2007? Doesn't seem very
> likely but then I don't know how good the Gnumeric XLS filters are so it
> might be true.
> 
> Ian



More information about the odf-discuss mailing list