Windows Vista Beta 2? Pass the Maalox, please
Summary: In about two weeks, Beta 2 of Windows Vista will be officially released to the public. And when it does, Microsoft will officially enter uncharted territory. I expect that Beta 2 will be reviewed as if it were a finished product. Those reviews will hit within days of its release, and they will be publicized more widely than any official Windows release ever. If you're a Microsoft product manager, it's time to stock up on antacids.
In about two weeks, Beta 2 of Windows Vista will be officially released to the public. And when it does, Microsoft will officially enter uncharted territory.
The last time the hoi polloi were allowed to tinker with an unfinished version of Windows was March of 2001. That, as it turned out, was an excellent time to slip under the radar. The once-powerful computer press had been ravaged by the puncturing of the tech bubble. A few pioneers were tinkering with crude tools for creating and managing blogs, but the big wave of blogging wouldn't happen until a couple years later. Newsgroups were strictly for hardcore techies. Not many people had the pipes or the patience to download a few hundred megabytes of OS code back then.
It's different today. Really different. The mainstream computer press has regrouped on the web, with advertising support and real budgets. Blogs allow anyone to express an opinion instantaneously, and thanks to sites like Digg and TechMeme, opinions get amplified and dispersed at lightning speed. And the target audience for Vista has more than enough bandwidth to handle its gigabyte-sized downloads.
This does not bode well for a complex, unfinished product like Vista. Dwight Silverman of the Houston Chronicle received a Microsoft-sanctioned copy of Windows Vista Build 5381 - billed as a preview of Beta 2 - [update 11-May 9:00PM PDT: Although Dwight's post says he installed "the latest beta," Dwight tells me he tried to install the February CTP release, Build 5308, which is the latest build Microsoft gave him to work with. Sorry for the confusion.] and his less-than-sanguine experience provides a preview of what you're likely to read a lot of in the next few weeks:
I've now got two -- that's right, TWO -- notebooks with partial Vista installs on them. Fortunately, the original Windows XP installation is still working on both.
On one notebook, an HP, the initial part of the install completed, but when the computer tried to boot to Vista to finish up, it burped to a blue screen with a stop-error message.
On the other, a Sony, the file installation never completed. It gave me a vague error, tried to roll back the installation and failed.
[...]
No wonder Microsoft put the brakes on the Vista train. At this point in the beta cycle, an installation should not fail catastrophically like this for two notebooks in a row.
My experience with 5381 was considerably more positive than Dwight's. It installed on my Tablet PC effortlessly, something I can't say about previous builds, and just about every piece of third-party hardware and software I've thrown at it has worked. The only real sticking point is a cantankerous connection to the free Wi-Fi at the hotel where I'm staying this week. I've been using Vista for months, and I'm impressed with the progress I see in this build.
But my experience is different from Dwight's. And most people who download and install Beta 2 are going to apply the same standards Dwight does. They will, reasonably, expect it to be polished and free of showstopper bugs. After years of using "betas" from Google and even from Microsoft, we've been conditioned to think of a beta release as the soft launch of a product rather than a snapshot of a work in progress.
When Windows Vista Beta 2 comes out in two weeks, I expect that it will be reviewed as if it were a finished product. Those reviews will hit within days of its release, and they will be publicized more widely than any official Windows release ever. I fully expect to read about bugs, crashes, installation errors, and even data loss - in short, all the things that we used to expect from beta software.
For the sake of comparison, I took a trip in the Wayback Machine to early 2001 to see what people were writing about Beta 2 of Windows XP. There was almost nothing on the web, save a few news stories and previews from the mainstream technical press. The newsgroups, meanwhile, were filled with reports of crashes, failed installs, and blue-screen-of-death errors. Back in 2001, most of those problems stayed under the radar. In 2006, Microsoft won't be so lucky. If you're a product manager at Microsoft, it's time to stock up on the Maalox.
Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.
Talkback
Any one that judges software based on a beta...
true
So, you are saying that there are NO standards for a beta whatsoever?????
So, if somebody starts trying to push pre-alpha code as beta, we should say nothing???? As long as they say "beta" we can't complain about Anything????
No, there are not.
As noted by the author, most installs go well. Obviously the problems reported by the other user is hardware related and finding those combinations making up the platform where the problems exist is part of the debugging process.
Consider how many possible variations of hardware there can be in the Windows world. This is not like other OS's supporting a select handful of components and/or drivers. (Or limited function drivers used by many in the *nix world.)
Anyone that signs up as a beta tester has to understand what they are agreeing to do, and that is to test *beta* code on their specific hardware.
Where do you draw the line on what is beta or not? I've always believed that call is best made by the person(s) creating the code.
It's still worth talking about
I agree it's wrong to pass final judgement on Vista's pending success (or failure) based on the beta, but it certainly is newsworthy knowing that the first beta has difficulties installing on certain hardware. It gives us something to look at for improvements on the second beta. If the second beta fixes the problems then that lessens the chance for delays, whereas if the problem persists that increases that chance.
Now if there is a delay I can live with that, because as I've said many times I'd rather it be right than on time. But if there is going to be a delay, I'd like to at least know about it, and people who test the software are usually much more forthcoming than the producers of the software when it looks like there will be a delay.
No there aren't
You can complain all you want (as you often do), but there are no standards for what any company wants to call a beta.
Windows XP and 2000 are still betas
CRAP.
Wrong...
The point of a beta is to be able to test and find the remaining glitches that only an expanded set of users would find. I.e. the ones that are beyond the capacity of the internal test groups to find, because they involve the many real world permutations of configuration and coexisting software that the internal lab cannot realistically reproduce. These remaining bugs should be more minor in nature, because they should be more configuration and application-related, rather than due to glitches in the OS functions themselves.
Since a beta is for testing usage in a wider set of scenarios, it must at least function under the major basic scenarios: it has to at least finish installation without dive-bombing. This is not happening, according to early reports.
Unpolished? I don't think so ...
RE: Windows Vista Beta 2? Pass the Maalox, please
NEVER Beta-Test for MS!
I was a Beta Tester for Windows Xp
I never expected a free or reduced copy of this software but since I did recieve it free so much the better.
I beta test for MS see the next os if I get somethng good if not still good.
Beta-Tester Blues (Continued)
You sound like a broken record!
Top people giving feedback came first.
I coudn't agree more.
how do I get it
http://www.arcon5.com
Windoze is ALWAYS a beta test
my goad
mabe a few bugs, but to make excusses is lame...
has anyone ever thought yhat the move to a 2007 release has some real reason why...
just lame
Spell check, anyone?