This article is more than 1 year old

Allchin lobbied Maritz to pull IE4 from Windows 98

Jimbo's heroic defence of integration is somewhat undermined by what he was saying during development

If Jim Allchin hadn't had the problem of exploding demos to deal with this week, he would have been determinedly singing the integration tune from the official Microsoft songsheet. But could this Jim Allchin by some chance be related to the Jim Allchin who, in March 1997, was arguing strongly that the integrated browser be jettisoned from Windows 98 development? Yes, we fear they are one and the same. In an email to Paul Maritz detailing subjects for a one-to-one discussion, Allchin says of the shell (yes, that shell): "Both the Memphis and NT team are totally frustrated with the IE4 situation: the code quality is lower than we can take, they are being driven to different objectives on when they have to fix bugs in Memphis or NT, the end-user experience isn't designed with migration in mind, our features we must have are not being done (this is a really bad situation), and finally, the code size/performance issues are serious. Given what I am seeing we are not on a path to have Memphis this year." Step back a little from that and take in the historical context. At the time, Microsoft has an announced strategy to integrate the IE4 browser into Memphis/Windows 98. It also has a convergence strategy (which you'll note got postponed again earlier this week, see MS junks consumer NT plan) for Windows and NT. Allchin's coders in March 1997 were facing precisely the same problems that have stalled convergence yet again. "They [the developers] say having two source trees is a nightmare and they have tried to check things in the IE4 environment and they are told they cannot because IE is trying to get a beta, etc. so progress here stops. The objectives are just different." Allchin presents Maritz with several options, one of which, "create a shell team here", he rejects immediately. Alternatively, they could "force a change of priority on the IE4 team". This appears to mean that the team should be forced to stop pushing for a beta of what would be a bit like, er, an application, and help the Memphis and NT people dig themselves out of their integration problems. But there are two other options he seems to like best. "Drop IE4 from Memphis and NT 5. There is a strong push to do this. We are wasting hundreds of people's time on builds that don't work, etc. Frankly, we may have to do this anyway to make progress. If we drop it, then we know we must either go out without IE4 in the final or we have to be honest in that both systems will take perhaps half year slip because we would have to fix the quality/performance/size later and go through beta tests much later." You can kind of tell he's attracted by this option, can't you? His other option actually relates to this, and is clearly something he's pushed for before, but been over-ruled. "Move the shell -- but not the browser -- to the OS team." That is, keep them separate, right? This is what Allchin has wanted for a while: "This was my recommendation before as you know. It may not be the thing you want to do for other reasons [yup, he was over-ruled], but it is the right thing to do for the OS (both Memphis and NT). IE4 would just plug into the environment [just like it did with Windows 95]. Both teams could make progress then. I still think it makes sense." And boy, does Jimbo love this option. The MS songsheet goes completely out the window, and he warbles: "We have to do something. It is not going to work the way things are today. I will be forced to do something this next week. It has dragged on too long. I must do something for the group as a whole to continue to make progress. If I had to make the call today the only thing I can do is remove IE4 from both systems and press on. At least then we can make progress [you said that already Jim]. I would have them move the win95 shell forward with the features we need for ZAW and simplicity." Email traffic from a couple of weeks later shows that Allchin didn't quite get his way, but that the case remained open. After some considerable debate over the pros and cons of integration, MS marketeer Jonathan Roberts closes the discussion with: "If it is apparent we will miss a December 1 ship, which is Joachim's [Joachim Kempin, head of OEM] cut off for spring machines, we have to make a trade-off decision between IE integration and hardware support. I suspect Billg will have to make that call since the implications are so massive." Microsoft did miss December, of course, so presumably Billg made the call, deciding the IE integration question once and for all for Windows 98. As recently as December 1997, do we presume? ® Complete Register trial coverage

More about

TIP US OFF

Send us news


Other stories you might like