Articles / Has Open Source become dera…

Has Open Source become derailed?

In today's editorial, Bruce Smith takes a "What have you done for me lately?" look at Open Source development, and shares his concern over the answer he finds. A few days ago, an article was published which asked whether Open Source was producing results. The author claimed that there were none, except for Mozilla. As a developer who as embraced the concept of Open Source, my first reaction was outrage. Obviously there were results, and many readers made the point clear. The BSDs, Linux, Apache, Bind, GIMP, and GNOME are but a few examples of highly successful projects.

But then I took another look at the article and at Open Source software in general, and I saw what he saw, and possibly the context that he saw it in. Until about a year or so ago, the mainstream press generally ignored Open Source and didn't really know what was going on in the field. Now everyone is jumping on the Linux bandwagon and the press are following the field for the first time. Look at Linux, Apache, Bind, and the rest. They were mature projects by the time the rest of the industry took notice. Since then, except for Mozilla, there have been no major Open Source projects that made a major impact on the market.

I guess I've just made myself serious flame-bait for all of you out there, but take a close look at the announcements on sites like freshmeat, SourceForge, or even LinuxBerg. For about the last eight months now, all I've seen in the line of new projects have been small utilities that do something that is often meaningless in the bigger picture. Sure, every one of us wants his ICQ client to look the way he wants, so those who can write their own and publish it for those who can't. I know I'm guilty of doing exactly that with my gView and gRun applications. They have been small tools I needed and that I decided the rest of the world might take an interest in.

I am not saying that those small projects should not have occurred, or that they are meaningless in themselves. My point is that these small projects are the only new items to come out of Open Source for the last while. Where are the killer applications, the Dream Weavers, the Visual Studios, the Exchanges, GroupWises or Lotus Notes, and the massive network management platforms? They are just not there, and while we don't do anything about it, the commercial programmers will either ignore the platform, or write their own versions and charge for them. And they will do it.

Maybe we should take a step backwards and look at the overall picture. There is something wrong with the image that we can see close up, like a puzzle where the pieces have fit, but are in the wrong place. Maybe it is a lack of seasoned developers and project leaders to create new projects, or a lack of programmers who feel that they have what it takes. I can tell you what it is not. It is not a lack of ideas or needs. There are plenty of those out there for us to take hold of and fill. We just need something to kick us in the butts and shove us into the deep end.

Right now, the rest of the world is looking at us with great expectations. We have made huge claims about how Open Source is the only way and that it produces better programs than the closed system. We know that is true. They see it as a claim, an unfulfilled claim on our part. We need to stop shouting to gain attention; we have it now. Now we need to start producing results to prove that Open Source is not just a handful of highly successful projects and a sandstorm of small, single utilities, or they will shrug their shoulders and pat Uncle Bill on the back again. And we don't need that.

So let's make a deal with ourselves: Let's pull the fingers out of the nether regions, climb down off the high horses, and get down to some serious work. I am working on a new project, been tinkering for a few months now, in fact. The project, tentatively entitled Rainfall, is a distributed backup system that can be used to perform backups on multiple systems from a single controlling console. It isn't as big a project as some, but it is something that I think we need. If you agree, contact me and we can try form a project team. If you disagree, contact me as well.

To Linus, Alan, Manish, Frederico, and all the rest of you that are out there: I give you a major vote of thanks and my eternal gratitude for what you have done. It has taken an lot of commitment, dedication, and endeavor from you to do what you have done. Now the rest of us must take up the rope and help pull as well. Otherwise, commercial software will win the war.

I invite comment, criticism, and anything else that you may feel is necessary to be said.

References


Bruce A. Smith <bruces@mail.petech.ac.za> is 24 and a 4th year IT student at the Port Elizabeth Technikon in Port Elizabeth, South Africa, with qualifications in chemistry and computer programming. He works as a general troubleshooter in Tech's Computer Services Department on a part-time basis. In short, a single white male with mild delusions of grandeur.


T-Shirts and Fame!

We're eager to find people interested in writing editorials on software-related topics. We're flexible on length, style, and topic, so long as you know what you're talking about and back up your opinions with facts. Anyone who writes an editorial gets a freshmeat t-shirt from ThinkGeek in addition to 15 minutes of fame. If you think you'd like to try your hand at it, let jeff.covey@freshmeat.net know what you'd like to write about.

Recent comments

28 Oct 2000 14:04 Avatar scottgnet

There are X domain spaces, and Y projects to fill them

There are 'domain' spaces which I define as an area of capabilty that an application addresses itself to.


Serving web content is a 'domain', and there are many packages that attempt to serve in that domain: Apache, AOLserver (my favorite) and many other, lesser known web servers. And now RedHat's TUX effort.


Once a domain has 'enough' applications that attempt to serve it, talent is drawn to those particular projects and there isn't usually anyone interested in starting yet another app to fill the same domain space.


Email servers is another domain. You mentioned there were no Open Source projects like Exchange -- we already have sendmail, qmail, postfix etc. that fulfill this domain area for us and do it well. But Exchange is 'much more' than that, and that may be the problem. Exchange tries to be everything to everybody -- they're tacking on workflow capabilities, they're tacking on directory services etc.; they're trying to integrate everything into it. But Unix and open source projects tend to steer for orthogonality, i.e. to be independent parts that do their job well, interoperating with other parts. Sendmail interacts with DNS et al. But they're separate projects and programs and you can replace DNS easily with another package that supports the same standards without breaking sendmail. With Exchange, they try to integrate everything into this monolithic program that does it all for you.


I'm not so sure that's a bad thing. Now a good thing would be to start a project that attempts to interoperate with MS Exchange, like Samba interoperates with Windows.

/s.

20 May 2000 14:46 Avatar astrobot

so what?
Oh no, we (the open source community) don't have any huge, bloated, ever-so-important, attention-getting "killer apps". What a big deal!

Arrgh! Everything we have is fine! A bunch of smaller apps, each extremely efficient in performing a single task, rather than a big app trying to perform every task. (I know that was kindof reduntant, but...) Strength in numbers is the key!! Strength in numbers is always victorious!!! Let us not conform to the standards and ideals that everyone wanting to be a part of open source wants us to accept. Why are people always constantly trying to change us and mold us? And why are we trying to take over the world?? We can do whatever works for us, and stay that way. If someone's mom doesn't know how to use linux and prefers a user-friendly OS, then let her use the user-friendly one, for Bob's sake!! Let us not mold to fit her requirements!!

If they don't like what we do, fuck 'em!

19 May 2000 09:16 Avatar y1000000

Building a very big town with very small Lego blocks?
[PS. I don't like writing in metaphores, but I hate to write opinions in technical terms. I'm learning ;) ]

I used to have a dream of building a town with very small blocks, so that everyone can visit it, walk thorugh it, admire it, and use it. And I'm sure I wasn't the only one. So why aren't there lots of big cities built of small blocks?

For a start, it takes time. A lot of time. Even a normal sized building would take ages, so it is almost impossible to build a town. So may be, MAY BE, we need some bigger blocks. If possible, instead of a very small blocks, I want to use buildings as building blocks. The problem is, of course, Lego doesn't make prefab buildings, because they want the maximum of variety. And that means it will take centuries to build a city. In the Open Source World (tm) we can fabricate bigger blocks from smaller blocks. And we should.

Secondly, there aren't enough kind of blocks. With Open Source, we have different kinds of prefab fire stations. One of them has lots of firehoses, another one has a hose and a ladder, and a third one has a car and a ladder. In the real world the third one can just borrow a hose from the first one, but with Open Source this is not done very often. Maybe because borrowing code is barred by rational reasons like "not done", "not invented here", or "I didn't know it was there".

Thirdly, there is no suitable "glue" for buildings in our Open Source World (OSW). In the real world, there are roads. But even that takes a lot of buildings blocks. In the OSW however, we would need something like Open Roads Between Applications (ORBA). Maybe it's just me, but ORBA is very little used between current applications. Most of things I've see are roads between a graphical front-end with an invisible application. Now I want to see more communications between graphical programs (and I don't mean graphical shells only). I really want to see a graphical application that can start and moves freely in a graphical text editor. (and if it is already there, please tell me, I want one for joe).

And lastly, other people want prefab cities. I've never heard of people who say they really wanted to change their city, except architects and construction workers. But that doesn't mean that there cannot be towns that can both be very rigid for its inhabitants and very flexible for builders, especially when everyone can have his/her own in our OSW. Having a town with a good school (a part of the core of a program) does not mean that that school cannot be copied to other cities (as a small core application in another application).

To recap, to build a Open Source World of cities, we need cities, roads, big building blocks, and small building blocks. At this moment (2000), we have some small building blocks (see the sourceforge projects), and a few cities (Linux, Apache for example). We still need some larger building blocks, but larger blocks can and should be build from smaller blocks. So, is it bad that everyone is focusing on the small blocks and the big cities?

[PS 2: I still use Delphi to build programs, because it allows me to use medium sized blocks (components), and small sized blocks (c/pascal/scripts) to build fabricated buildings(small generic applications). In future, I do want to use something like that for Linux (can someone help me with this? I don't know what to look for).]


--

18 May 2000 11:42 Avatar alexbm

I agree
I have just read your article, and this is not a flame. In fact, I wholeheartedly agree.

Development for Open Source has almost ground to a halt. Development for Linux is even worse, with 3/4 functional USB only just becoming available in the latest kernel. Plug and play is still sketchy, and very little interfacing with todays commercial leaders.

Where is the mail client with an Exchange Client plugin? I find it hard to believe that Microsoft would deny this privelidge, mainly for the bad press they'd get. What about a half decent browser? Netscape is slow and bulky at the best of times, Lynx not cutting it with todays web sites. Opera would seem to be a hope but has still not made it out of alpha test.

Open Source has the potential to obliterate the commercial variants with the stability and power behind the apps and OSes. The developers need to get their finger out, and forget the &quot;Open Source is for experts and needs to be kept complicated&quot; attitude. Make Linux and related products easier to use, and its bye bye Microsoft.

I tell you, the only reason I still use Microsoft products for example, is they do the things I want them to do, easily. Do that on a free operating system for me, and make it legal for me to have it free, then its bye bye MS. In fact, my main OS is now Linux, with vmware loaded for those things I can't do yet.

I do hope to be getting rid of vmware soon ;)

09 May 2000 05:16 Avatar spodzone

Size and shape of Big Picture(TM)
I think it's a question of expectation as well; either you're interested in following the commercial world where you have millions of companies all spewing out code with stuff-all interest in it, and/OR you're interested in the community that codes for pleasure to fulfill a need - opensource and the FSF strike me as more about writing something small and neat to get the job done (and &quot;maybe someone else will benefit from this&quot;) than the former.

Screenshot

Project Spotlight

Kigo Video Converter Ultimate for Mac

A tool for converting and editing videos.

Screenshot

Project Spotlight

Kid3

An efficient tagger for MP3, Ogg/Vorbis, and FLAC files.