McVoy Strikes Back

cranos writes "Fast on the heels of his previous article claiming the kernel is at risk of Bad Things over the BitKeeper fuss, Daniel Lyons has released a new article where Larry McVoy attacks the Open Source movement as non-innovative and dependent on the kindness of corporations. The following quote says it all: 'The open source guys can scrape together enough resources to reverse engineer stuff. That's easy. It's way cheaper to reverse engineer something than to create something new. But if the world goes to 100% open source, innovation goes to zero. The open source guys hate it when I say this, but it's true.'"

Re:McVoy doesn't get it

[ssj_195]

To begin with, software these days is quite complex and it really is impossible to have a full-blown operating system with all the applications people expect and not have some sort of issues. Secondly, the vast majority of people out there are not computer savvy and are going to need help regardless of how well built their OS/applications are. Red Hat isn't dead yet so I wouldn't be so quick to proclaim them as such, although their demise wouldn't entirely surprise me.

Indeed. Non-trivial software will require support (either to install, or to tailor to your companies specific requirements) until we invent Strong AI, not before.

That's one of the great things about open source software; it doesn't have to. Companies like Red Hat are packagers, not necessarily creators. What they provide is a nice, neat package of what others are already creating.

Even then, Redhat to a *huge* amount of development, especially on GCC. If I recall, Luminocity was also funded by Redhat. This is not even close to an exhaustive list.

It wasn't even reverse engineering the program

[Lifewish]

If I recall correctly, all that was "reverse engineered" was the client-server protocols. This is the same sort of thing that the EU is currently yelling at Microsoft to release to the world, as keeping it quiet is a great way to lock people in to a product.

Re:I can't disagree

[MrDomino]

Sure most have one or two innovative features, but what applications in the OS world are really innovative, especially from an end user perspective?

Certainly not desktop [enlightenment.org] environments [gnome.org], servers [apache.org], remote shells [openssh.com], anonymizing [freenetproject.org] (or swarming [bittorrent.com]) networks [eff.org], or compilers [gnu.org].

Because all of those things are just replacements for commercial applications, and did nothing new.

Re:So, you programmers ready to give up your jobs?

[Tim C]

But if it's GPLed, you can't prevent your customers from undercutting you. The effect of the GPL is to push the price of the software to zero - you can only realistically make money selling related services, such as support or bespoke modification.

Re:Non-innovative?

[gclef]

ummm....not that I'm disagreeing with you, but McVoy didn't write Subversion. SVN's an open-source project much like Arch, Monotone, etc. McVoy did Bitkeeper.

(just doing fact-checking....we now return you to your normally scheduled slashdot discussion)

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Well, let's have a look

[Daniel Dvorkin]

PHP? Yet another clone of MS's ASP. Yes, MS did invent that kind of server-side inline scripting. (Yes, I know they're supposed to never have invented anything. Sorry 'bout letting reality get in the way of that.)

Reality: the original PHP (PHP/FI) was developed in 1994, released in 1995; ASP was released in 1996. Sorry to shatter your precious illusions.

Re:Well, let's have a look

[cheesybagel]

Mozilla comes from Netscape. IE came because of Netscape. They were both either clones or directly based on Mosaic. Which was made by academia (you know, NSF funding, which some people say isn't useful for anything).

Several things had plugins before Active X. Cubase has had plugins for yonks.

Microsoft Office started with Word and Excel. Word is a Wordperfect clone and Excel a Lotus 1-2-3 clone.

FTP was also file sharing before Napster. It just wasn't p2p. p2p apps predate Napster (e.g. the military has used them for a long time for massive distributed networks). Napster however, sucked.

Napster was basically IRC with DCC and a search engine tacked on top. Hardly innovative, and I still remember how it was lame at resuming broken downloads, how it didn't segment downloads and did not check if files had errors in it.

BitTorrent on the other hand segments downloads, does proper checking, and works well for downloading large files. Napster didn't. Napster was for mp3 files, and sharing anything else was broken.

Before PHP and ASP there was cgi-bin, which you could write in shell script, which predates either of those by a long time.

Everyone and their grandmother has done SQL after IBM. Including large closed-source companies like Oracle. Does not mean theirs aren't better in some way of course. Just because you made the first version does not mean your version will always be better.

Besides these are non-sequitor, because I didn't see the parent poster mention them. You just threw them in to prove your point. If anything you just proved closed-source commercial works are as big a bunch of cloners as opensource, if not more.

It's just a standard response to Freedom.

[btarval]

What's also interesting is that McVoy's response here is the same exact response that closed-source vendors ALWAYS make when Open Source starts costing them market share. McVoy's statements are nothing new; just a variation on a them.

Let's see some examples:

Microsoft. The OS, Webserver and IE are all classic examples.Their attacks on Open Source are in a league by themselves, including the "stifle innovation" argument of McVoys'.

Windriver. These folks bashed Linux mercilessly while their marketshare dropped from 35% in 2000 to 14% today. They threw in the towel and went with Linux last year (though VxWorks is still around, it's clearly not the priority).

GreeenHills. These folks have been bashing gcc for years, as the embedded market has moved away from speciality development tools except in certain small areas where the performance is required.

So McVoy's response is nothing new here. He must be feeling the pinch of people moving away from his software.

Now, if Slashdot would only stop giving him free publicity, we'd be all set. McVoy has already stated that everytime he's mentioned on Slashdot, his "sales go up".

Re:McVoy doesn't get it

[Anonymous Coward]

Actually, that would be a hypocrite.

I wouldn't lose mine

[Dog135]

In every programming job I've had, there didn't exist any tools that did the job properly. Most of which was reports.

I remember one time, the managers tried using a reporting tool they bought to make a daily report. Unfortunately, it took 26 hours to run. After one of the programmers rewrote the report by hand, it ran in under 2 hours.

And there's lots of web development that can't be done with webpage writting programs. I wrote lots of serverside scripts at my last job.

General purpose office applications are a small niche market in the sea of software development. The only people who'll loose their jobs are those working for MS.

Re:Well, let's have a look

[Alioth]

Well, firstly, file sharing was NOT pioneered by Napster. We were using IRC to file share peer-to-peer in 1990, probably years before the writer of Napster had heard of the internet. In any case, saying BitTorrent is a copy of Napster is so wrong it's not even wrong. It works in a completely different way (and is designed to solve a different problem). The only similarity with Napster is it allows peers to exchange data.

PHP came out before ASP too. You are not letting reality get in the way of anything, because Microsoft did not invent server side scripting first.

In any case, the first web browser was open source. The first web server was open source. The first TCP/IP stack was open source. The first SSH was open source. The first network transparent windowing system was open source. There is no closed-source equivalent of rsync.

McVoy is bullshitting I'm afraid.

Re:wouldn't need to

[ArmpitMan]

Psst -- "telbij" and "vrmlguy" are probably different people.

Re:Well, let's have a look

[Wdomburg]

The C implementation of PHP was released in 1995. Language constructs existed in the beta versions also released later that year.

Whether you think its a "shoddy piece of work" or not, it clearly isn't a clone of a product released a year later.

Re:Counter examples

[radish]

Agreed. Firefox is a great product, and I use it all the time, but innovative it is not. I can't think of a single significant feature that is new in Firefox (tabs, popup blocker, extensions, skins/themes, standards compliance, etc). What is new, and very significant, is the assembly of all those great ideas into a single product. The fact that it is Free is the icing on the cake.

Re:wouldn't need to

[Mr. Slippery]

I didn't say I wanted it to be GPL, nor did I say that I wanted it to be "free". You said it was "open-source", so I asked where it was.

Software that is GPLed is, by defintion, open source. [opensource.org]

There is no mandate under the GPL to release the sotware to the public or make it generally available, only to make the source open to those to whom you distribute binaries.

By your "where is it?" question, you seem to be assuming that open source software must be publically available. This is not the case.

Re:So, you programmers ready to give up your jobs?

[Jherek Carnelian]

This - in my view - is one of the deficiancies of capitalism. It takes the eyes off what was supposed to be the ball, and instead force focus to profits

It isn't a deficiency of capitalism per se, but rather the flawed implementation of it that we currently have. There are two reasons Ford can be successful profitwise by making poor quality cars:

1) The market values cheaper priced cars over higer-quality cars - that is good capitalism because the market demand is being satisified. You may not agree with that demand, but it will continue to exist regardless and attempting to deny that demand would be an expensive, long-term loser.

2) The market has been manipulated through non-free-market forces such as government regulation (protectionism and the like) or information-hiding (like lawsuit settlements with non-disclosure requirements). That is capitalism diluted with corporate welfare and is completely bad and should not be allowed to happen in a true free-market.

Re:McVoy doesn't get it

[winwar]

Well, in some sense the services model IS based on the idea that you are giving your customers crap. It is how you define that word crap. For instance, does it mean something that works poorly or something that doesn't do what the user wants even though it was advertised to do it. In either case, from a user point of view, it sucks, especially if you have to pay to just get it to work. Great software should be easy to configure and do what is wanted/promised. Granted, it's not limited to open source.... Definitions matter.

Finally, what exactly is the "services" model? I mean, I would consider any software that the service contract costs more than the price of the software a services model.... I suspect that would include Bitkeeper.

Re:McVoy doesn't get it

[toby]

in some sense the services model IS based on the idea that you are giving your customers crap. It is how you define that word crap. For instance, does it mean something that works poorly or something that doesn't do what the user wants even though it was advertised to do it. In either case, from a user point of view, it sucks, especially if you have to pay to just get it to work. Great software should be easy to configure and do what is wanted/promised. Granted, it's not limited to open source.

I agree with most of that. But Larry's wrong to slur open source as "shipping crap". He knows perfectly well that paid-for and shrinkwrapped software is no better and often worse. It's a transparently dishonest slur meant, of course, to give proprietary developers such as himself a superior glow. In other words he's slinging the same old inglorious muck: FUD.

In my earlier posts I did gloss over the difference between "fixing stuff that's already paid for so it works as advertised" and "adding new stuff that I want". It's interesting to compare the implications of these two needs in the proprietary and open source spheres.

In the proprietary world, many companies will provide free fixes to features that users already paid for, as a matter of policy or principle. (Many users expect this, being accustomed to warranties on tangible goods - "why should paid-for software be any different?" Despite EULAs I imagine there is still even some statutory protection.) Examples include Apple and even Micro$oft. However, as I mentioned, this is becoming less true; Adobe is a notable counterexample. By and large they no longer distribute bug fixes, so if you bought a broken feature, you are screwed and must pay for the next release and pray for a fix. This attitude ("we're not going to make it work as advertised") creates tremendous ill-will and is (I hope) self-defeating in the marketplace.

In the free world, if a feature is broken, the professional pride of the developer normally results in a quick fix. In those two respects, the two spheres are similar; policy or attitude dictates the response to "broken features". The exceptions might be where resources (funding) is not available, and the developer must wait for support or sponsorship in order to fix existing features.

When it comes to adding functionality or initiating a new project, of course, the proprietary world invariably exacts payment. This is true, for instance, when Apple does a new major release of OS X, or when Adobe revs Photoshop. In the free world, new stuff appears magically if the developer is motivated (maybe scratching an itch of their own), or if the developer is unresourced, it appears when sponsorship appears (maybe their employer has an interest in the feature, and subsidises development).

I see the "services" model as referring to "packaged" agreements which cover developer sweat on behalf of a customer. Depending on the package in question, that could include tech support of an educative nature ("how do I do this?"), bug fixes ("this doesn't work") and even additions ("I need this"). In my case - and that of many other free software developers - I usually provide all these with no expectation of payment, since I care deeply about reputation, and helping customers.

Certainly Larry seems to have a crazily warped view. I don't see what is so hard to grasp, or far-fetched, in the above. Having such a phobia of service-oriented business, unless he meets RMS on the road to Damascus, it's doubtful he'll ever stop spreading FUD and join us. One might at least hope that the media does a little more credibility checking before they blindly quote self-promoters like him. (As if!)