New Sidekick Will Run NetBSD, Not Windows CE 262
jschauma writes "Many sites are reporting that the next Sidekick LX 2009/Blade, from Danger (acquired by Microsoft early in 2008), is going to run NetBSD as their operating system, causing Microsoft's recruiters to look for NetBSD developers."
Re:Embrace. (Score:3, Interesting)
No, this is notable because it's an open admission that WinCE can't cut it .
At least in the short term. MSFT appear to have bought this product from elsewhere. To keep it alive they need to get a release out the door. Maybe in parallel they are porting the software to run on WinCE.
Even better... (Score:3, Interesting)
This is the Postfix program at host mailxxx-xxx-R.bigfish.com.
I'm sorry to have to inform you that your message could not be delivered to one or more recipients. It's attached below.
For further assistance, please send mail to
If you do so, please include this problem report. You can delete your own text from the attached returned message.
The Postfix program
: host xxxxx-xxxx-mail5.customer.frontbridge.com[131.107.115.214] said: 550 5.7.1
$whois frontbridge.com,
Domain Name: FRONTBRIDGE.COM Registrar of Record: Corporate Domains, Inc. Administrative Contact: Microsoft Corporation Domain Administrator One Microsoft Way Redmond, WA 98052 US domains@microsoft.com +1.4258828080 Fax: +1.4259367329
$whois bigfish.com ,
Domain Name: BIGFISH.COM Registrar of Record: Corporate Domains, Inc. Administrative Contact: Microsoft Corporation Domain Administrator One Microsoft Way Redmond, WA 98052 US domains@microsoft.com +1.4258828080 Fax: +1.4259367329
Re:I never thought I'd see the day. (Score:3, Interesting)
Re:Even better... (Score:2, Interesting)
I once worked with an Indian consultancy company. They were working on software for a mobile phone and were using ClearCase. Now ClearCase is expensive, but it does the job. A colleague and I were writing code to test the peripherals on the baseband chips. Now one of the Indian managers said that the project was stalled because the I2S controller didn't support some mode.
Since we had code that tested the I2S controller, we were drafted in to help them. We asked for a config spec. It looked like this
foo.c@\main\1212
bar.c@\main\1254
and so on, for thousands of files. Even worse if you emailed a few different people, you'd get slightly different config specs, but always of this form
Now normally in Clearcase you develop on a branch and then merge to a release branch. So the development config spec will be something like this
element * CHECKEDOUT .../developers_branch/LATEST
element *
element -file * RELEASE_LABEL_1 -mkbranch developers_branch
What this means is take the checked out file if if exists, otherwise look for the one on my branch, otherwise look for the released version
and a release one will be like this. Once you're done developing you merge your branch back and label the result with a new release label. Then the config spec looks like this.
element -file * RELEASED_LABEL_2 -nocheckout
Of course for this to happen you need to have a management structure that makes sure people get things right before they merge them back, and if two teams of people are fighting that things get resolved. Otherwise you end up with a minority report situation where different bits of the team end up working on different baselines.
The worst case of this would be where everyone picked a set of last known good versions that were different. Which is exactly the situation these guys were in.
Now you can see that the spec they sent us showed that something was very wrong.
We managed to get it to build but some things didn't work, actually the things we wanted to check. So we asked them and they said something like "ask Raj, he's go a fix for that". The fix was one file, which he emailed you. You checked the file out and overwrote it with the one in the email.
At this point, it was clear that the I2S settings were totally wrong. We fixed those and managed to punt the whole thing back. Given the chaos the project was in, I didn't really expect it to ever work properly.
It was the most amazing misuse of a version control system I have ever seen. What was odd about it was the individual developers seemed to me to be ok, the problem was the shitty consultancy company was loading them down with work without setting up things like version control properly. Actually I always suspected that the project we saw had been put together in a few hours by some very smart people, who had then billed my client for a shitload of hours which hadn't been worked. Then after that they handed over the whole mess to some much less experienced developers who were basically too timid to realise that they needed to do a drastic set of module tests, merges, system tests, bug fixes and so on until they had a stable baseline to work from, because the alternative would be that the project would crash and burn.
Still I'd never trust one of those big Indian outsourcing companies to do software after that. And as I said, it's a problem of the company, not the developers. With one decent manager, the project I saw would probably have not got to this dire state. Actually with one decent manager they could probably have pulled themselves back from the brink given a month or so.
Re:Just a minor note (Score:3, Interesting)
It's actually the other way around - they used to show up in Netcraft as Linux servers even though they were IIS on Windows Server 2003 for a long time.
This is because the server version reported was actually Akamai's balancing and caching infrastructure in front of the Hotmail servers.
What about the Microsoft Xenix Sale Agreements? (Score:4, Interesting)
Re:Embrace. (Score:1, Interesting)
The problem is not that the code is used and locked away from the original developers. The problem is how the GPL crowd is acting to those original developers. The major problems are:
- GPL fanatics always talk about slavery and morality and stuff, but act just like those they pretend to oppose.
- Often enough the companies contribute to BSD projects because they know it is cheaper for them to do that instead of constant merging. The GNU just wants to own it all because it is their moral obligation.
- The misconception of most GPL developers that you can do anything to BSD licensed code, including releasing it under the GPL license. That is not true, you are not allowed to replace the license. You are allowed to release the combined work under any license you want, but if you release the source the original code still must be BSD licensed.
So in general the problem is that the loud GPL crowd is seen as fundamentalists by the BSD crowd. And we don't like fundamentalists. Sometimes it feels like they only add GPL code to BSD projects because they want to prove their point, making time and effort for everyone higher without any real benefit other than stimulating their egos.
Re:Embrace. (Score:4, Interesting)
Actually, for the longest time, MS was going to move the Sidekick over to WinCE- they were even gearing up for it. Unfortunately, after many months of this (A year ago, in reality...), they have announced that they're doing it with a *BSD core and they're HIRING *BSD devs for it.
If you're doing what you're claiming, you don't spend 12 months doing it that way and then gear up for the other OS that you don't sell...doesn't look good to investors to spend 225 billion or so on someone to do something like this. ;-)
Re:What about the Microsoft Xenix Sale Agreements? (Score:3, Interesting)
They're not selling the OS. They're selling the phones which use an OS.
Doesn't breach their non-competes with SCO, sorry.
Re:Embrace. (Score:1, Interesting)
Rubbish. Apple can create a GCC-based syntax highlighter that communicates with XCode through sockets. Licensing problem solved.
Actually, linking two programs via sockets is a legal gray area with the GPL, and I'm not sure it's even allowed in GPL 3.0.
Note that this wouldn't apply to a Internet-based server communicating via open standards, where any conforming client could connect, but it would apply to two programs coupled together so tightly that they can't operate without each other, in which case the socket is considered linking, and thus the GPL applies.