Open Source is 'Not Reliable or Dependable'

Exter-C writes "News.com is reporting that Jonathan Murray, the vice president and chief technology officer of Microsoft Europe has made claims that 'some people want to use community-based software, and they get value out of sharing with other people in the community. Other people want the reliability and the dependability that comes from a commercial software model.'"

Comments

*boggle*

[Akardam]

Microsoft is one to make claims of reliability and dependability.

Re:*boggle*

[Dr. Evil]

We can rely that there will be security updates and we can depend upon them utterly.

So it's a reliable and dependable model.

Re:*boggle*

[mausmalone]

Since switching to Win XP from Windows 2000 during RC1, I've experienced a few crashes due to some bad ram, but beyond that it's been steady as a rock. Also, I haven't needed to waste resources with a virus checker because I know how not to get viruses. A good firewall goes a long way.

Here's the question you have to ask yourself, though... will your friends and relatives who don't use OSS and who have crashes & viruses actually do better with OSS and a fresh install of Linux? Or would their problems be fixed with a fresh install of Windows, a good firewall, and the abolition of Internet Explorer?

I think that if most Windows users just used to use Windows in a safe way (and read the fucking dialog boxes that came up instead of reflexively clicking "OK" to everything), a lot of the "unreliable" and "virus-laden" views of it would start to dissipate.

While I know that Linux and OSS can be very secure and stable, Windows can be also. If people put the time into Windows that Linux-users put into Linux/OSS (by way of customization, and finding apps and drivers), they'd have a much more reliable machine (than their current Windows install ... I have no desire to compare Windows and Linux). The biggest unreliability with Windows is the stupid things that users do.

Re:*boggle*

[rthille]

You recommend abolishing IE, but what other commercial web browser is there for Windows? I guess there is Opera, though I don't think it has much mind-share...

Re:*boggle*

[Moofie]

"The biggest unreliability with Windows is the stupid things that users do."

The biggest unreliability with Windows is the reliance on users not doing stupid things in order to stay reliable.

That's a dumb thing to rely on. ASSUME the user is going to do unwise things, and design around that assumption.

Re:*boggle*

[ahodgson]

But the article says open source is not reliable. Ergo, no firefox.

Re:*boggle*

[ncc74656]

You recommend abolishing IE, but what other commercial web browser is there for Windows?

Why does your browser have to be a closed-source product? Last time I checked, Firefox runs pretty nicely on Windows. If anything, open-source apps running on Windows can serve as a bridge to eventually running open-source apps on something other than Windows. If a file created under (for instance) OpenOffice on Windows opens without issue under OpenOffice on Linux, that's one less impediment to eventually switching away from Windows.

Automatize please

[Opportunist]

Could we simply auto-tag all stories containing "Jonathan Murray said" as "fud"? It would save a lot of work and I doubt we'll get too many false positives.

Nice FUDdy title

[xtracto]

how did slashdot editors managed to understand "ther people want the reliability and the dependability that comes from a commercial software model.'"" to "OpenSource is unreliable".

Hey, sometimes Open Source does it right, someties people preffer other ways. If THERE ARE companies that sell CLOSED software and services and their services al GREAT, yes this is FUD, but this time it is the editors the ones that are throwing it.

BURN KARMA BUUUUURN!!

More self-serving propaganda-- is this news??

[postbigbang]

And really now, what did you expect him to say? Our model sucks, and please, let me now genuflect in the hotbed of OSS dev?

It's like asking Steve Ballmer to take estrogen.

Not dependable?

[DaveM753]

I would have replied to this sooner, but Windows keeps crashing.

*shrug*

[stlhawkeye]

I don't think it's any more reliable or dependable than any other development paradigm. The difference is that instead of paying somebody for unreliable and undependable software, I can get it for free from open source. Firefox crashes more often, on every environment on which I run it (4 different OS's) than any other application I have. The difference is, I didn't have to pay for it.

Re:*shrug*

[Arker]

Another difference is that you can, if you wish, actually help make it stable!

Well, in firefox's case that would probably mean forking it since the development team has a chronic case of featuritis, but again, you can do that if it's important enough to you.

There are some definite advantages in terms of reliability and security to the free software model, but that doesn't mean all free software is going to be more reliable or more secure than all proprietary software - far from it. Free software, however, does allow users to become involved and part of the process, rather than condemning them to exist only as passive 'consumers.' And it does respond to their needs, rather than to the desires of the marketing department.

IE is much better coded than firefox - and firefox therefore crashes more often. Yet, despite that advantage, IE is much less *secure.* And that's what you get when marketing determines the program specifications...

SourceSafe vs CVS

[Rik Sweeney]

I bet to differ Microsoft. Why would I use SourceSafe, which is slow (checking out takes a very long time), unreliable (corrupts itself regularly) and costs money when I can use CVS which is fast, reliable and is free?

Re:SourceSafe vs CVS

[955301]

Just a few questions.

First about "I tried to install CVS or subversion". So, which one was it?

Second, you seem to value speed on something you do once - installation and setup - over the steady-state use of the source control tool - keeping your data integrity intact.

For Subversion, the explorer client is TortoiseSVN.

I've used Source Safe, Clear Case, Starteam, CVS and Subversion, RCCS, and a few others I've forgotten. By far, Subversion has been the best. Starteam was close, but it required a Microsoft setup back when I used it.

I do not agree that Subversion is difficult to set up.

Re:SourceSafe vs CVS

[hyfe]

Comparing SourceSafe with CVS is like comparing MS Office with Emacs ...

.. it's inheretly flawed because anybody remotely sane will recognize that SVN and VIM are superiour products.

Strange...

[biglig2]

...I mean, why are they so freaked out by Google? Since their entire infrastructure is based around software that isn't reliable or dependable, they can't possibly grow to any size.

NEWSFLASH!

[rolfwind]

Microsoft Executive will try to talk you into buying commercial software! GASP!

Well... actually, he said "commercial", so perhaps he's suggesting Mac OSX:) Perhaps he can clarify if he's trolling for his own company's software or if he means all commercial software. In which case he's not a marketing troll, but an idiot using a blanket statement who clearly doesn't care about the issue as he should be aware that Microsoft has used Open Source components in it's own OS - (TCP/IP stack?) - whereas they could have used a "superior" commercial solution.

Re:Marketing tripe

[xtracto]

Of course people want dependable, that's why they're looking for something not laden with spyware, viruses, etc.

Sorry, you did not got FP.

I really hate the slashdotters that have this logic "ClosedSource -> Malware" or "ClosedSource->Bad", there are tons of applications that are closed source and DO NOT have any kind of crapware on them, a lot of them are even FREE.

Just because the author of a program do not want to give you his lunch for free does makes him baaaad, anti OpenSource or whatever, come on, get a grip!

MicroJerk!

[digitaldc]

First Microsoft flirts with Open Source saying it's 'maturing and more commercial,' and now they say it is not 'reliable or dependable.'

I think they are just badmouthing them because Open Source won't let Microsoft go all the way on the first date.

RTFT

[mechanosm]

That's the title of the article at news.com.com.com.com. Perhaps you can address your question to the editors there.

Where would you live?

[DumbSwede]

So to make an analogy, I should prefer buildings that are built that allow no inspections while being built or even after construction is completed, to buildings that are free to be inspected. Which would you trust to live in?

I have to agree in general

[erroneus]

Because, after all, what is "reliable" or "dependable"? By whose standards?

I just loaded FC5 on a machine cleanly. I then had it do a yum update. Once completed, firefox was unable to start as a regular user. (Root could start it.) Turns out that somehow the ${HOME}/.mozilla directory was chown root.root for some reason. I changed it and all was well again.

So yeah, it's "imperfect."

But GOOD-FREAKING-GOD! This is Microsoft claiming this? As if they set the standard for reliability and dependability? All this while their EULA states that their software is not guaranteed to be suitable for any purpose at all. That just OOZES customer-service, reliability and dependability.

Ridiculous...

Well of course

[sterno]

It would be rather a strange thing for a company totally dependent on the sales of proprietary closed source software to go out and talk up how wonderful open source is. It would be similarly looney to expect say, RMS, to talk about the advantages of closed source software. News for Nerds: Stuff that's obvious.

Hmmm

[linvir]

While I'm not an IT expert, so I can't say anything particularly clever, there is one difference even a lowly dickhead such as myself can see.

When someone stops supporting an Open Source product, it's still available to be updated by the community. When Microsoft decides that it's time for you to buy the latest version of their OS, you have NO FUCKING CHOICE. That's not dependability.

There are inherent problems with closed source!

[postbigbang]

Let's review:

1) collaborative third party development and evolution is impossible with closed source, except by a proprietary gatekeeper of some type
2) visible source is easier to fix than invisible source
3) it's impossible to judge application quality and security without seeing source; otherwise it's hearsay
4) open source survives the ills of its progenitors
5) it's still ok to charge for software, even open source, IMHO
6) trade secrets can be encumbered by closed source, and so can lots of copyrights and patents not owned or licensed by its developers
7) you don't learn by reading closed source code (an oxymoron), however, you can learn by reading open source code
8) closed source doesn't actually suck, but it can be used to hide, obfuscate, cajole, and frustrate both developers and users

OS/2 was a technical success and market failure, and took eons to get bug fixes finished. The same can be said for BeOs. Simply building a better mouse trap and thinking that people will flock to you is one of those sweet lies that duped engineers believe. It's simply not so.

And now Apple probably sucks because their microkernel and some of their codebase is now closed. For that, we'll all suffer.

No Kidding

[Greyfox]

I maintain a program that runs builds on pretty much all the commercial UNIX and Windows platforms ever. I have a minion who devotes a couple of hours a day to unsticking and requesting reboots on Windows systems that have gone down during the night. The Win64 machines are particularly bad -- one or two of our 6 machine clusters BSOD daily. It's random as to which one goes, but we've run memory testing on all those machines and they check out fine.

UNIX machines, including 32 and 64 bit versions of Linux go down infrequently enough that I investigate personally when it happens. We've had two hardware-related cases of UNIX machines becoming unresponsive to telnet and ssh requests in the past 6 months or so.

Reliability. Hah. Like how Outlook likes to remind me 7 hours after a meeting that I'm 7 hours late for the meeting. It couldn't be bothered to let me know before the meeting, mind you. That would be too convenient.

Microsoft has no clue what reliablity means. Some marketroid in Microsoft shouldn't be shooting his mouth off about how reliable their software is, when he's obviously never used reliable software. I'd like to address the following personally to the pencil pusher Jonathan Murray: "Shut the fuck up and go back to trying to convince companies to drink your company's poison kool aid. I dream of the day when your products are so marginalized that I never have to use them ever again."

Re:BULLSHIT

[linvir]

Just because Microsoft stops officially supporting a product does not mean everyone has to run out and get the latest version

WRONG. You're one person. Arguments based on "you're full of shit because your point doesn't apply to me" tend not to work. There is a wider world of people out there who need security updates and other patches. Most important of all of these are the business and school networks, Microsoft's real source of money. They have to stay up to date, or they get owned.

Now, the real issue is whether or not the updates are the source of the exploits. If MS didn't reveal the flaws, maybe there wouldn't be so many exploits for the unpatched systems. You might have had an interesting post if you'd gone with this, instead of two long paragraphs of narcissistic swearing. Do you understand that, dickhead? I don't give one rat's ass how you use your computer, I'm using my vague knowledge of IT in general. Let me clue you in on something: YOU ARE NOT THE YARDSTICK BY WHICH THE REST OF THE WORLD IS TO BE MEASURED.

Yea, and green bikes are faster.

[the_REAL_sam]

"Open Source" means that the source code is visible to the public.

Even licen$ed $oftware could be open source.

It is really quite silly to base generalizations on software's reliability upon whether or not its source code is visible. It's tantamount to saying "green bikes are faster."

On the other hand, the reason open source software is desirable is that it fosters trust on the part of the user. When I say trust I mean that the user can look into the source code of the software, and verify that it:
opens no backdoors,
installs no rootkits,
does not locally snoop,
does not locally spy, spam or advertise,
or leech system resources,
or delete the user's files,
or mess with security levels,
or alter files that it doesn't own,
or send out a flood of packets /ddos,
or hack remote systems by means of worm or proxy,
or open a local port,
or port scan and relay,
or be a blockscanner,
or a wardialer,
or do any of those other nasty things that we've seen and/or heard of.

in other words, open source software helps the user to verify that the executable software it compiles will not hack remote systems, and will not hack the local machine, either.

that's not to say i know anybody that sits down and reads the open source, any more than i know anybody that reads the full license agreement before clicking "i agree". but "trust", that's the theory.

there's also the creative commons aspect of it, as in "the software engineer you help train to day might be the one you hire tomorrow." if the guts of the software are visible then others can learn and share, and build upon each other, providing the best overall source code.

i've heard arguments that such a thing opens the door to piracy or software plagerism, risking profit loss. Well. Consider how many HUMAN hours went into writing and re-writing the same code based on some business man's notion of profit. Jesus Christ said that the love of money is the root of all evil.

An anoying AC asks for references.

[twitter]

The vast majority of internet-facing computers that function as zombies are in that state thanks to user intervention. I dare you to prove otherwise. Because if that wasn't the case then every single "Windoze" computer on the internet would be a zombie, and that is not the case, now is it?

Well, the majority of M$ computers ARE infected. It does not take long and it requires no "stupid" action by the user. Indeed, no action is required other than plugging the thing in. Study after study has shown this, but here are two for you:

• 2002, BBC, most read english language news site, tells users to "avoid Internet Explorer" based on drive by exploits.
• 2004, Botnet responsible for 80% of world's spam
• The Internet Storm Center, 2004, 20 minute half life.
• Avantgarde Study, 2004, 4 minute minimum time.
• A 2005 study of zombies, a total underestimation but one that shows large numbers.

Things have gotten worse not better and the numbers match personal experience all of us have. I've seen people bringing broken computers into stores. I've seen broken computers in banks, you know, the ones so far gone nothing can be done. While a user can help the process by going to net nasty sites, it's still not the user's fault. Their computer should not fail them that way.