in Search

Do you Twitter? Follow us @CodeBetter

Eric Wise

Business & .NET

Is Open Source Killing The IDE/Tools Market?

So, I see that Delphi is for sale. I'm actually not all that surprised since there are free alternative IDEs popping up everywhere, the most notable of which is Eclipse. The IDE and tools market seems to be losing margins rapidly as free components and tool sets compete with traditional corporate offerings.

Is this a good thing or a bad thing? I often note that the Open Source community seems to play catch up to the for profit offerings. Open Office for example has been chasing Microsoft Office for years and each time it starts to catch up Microsoft has to add new features and functionality to remain dominent. In a way, this is a good thing because it keeps Microsoft from being lazy but in a way I can also see this being a bad thing if eventually the free offering gets "good enough" that Microsoft would decide to leave the market. Why do I feel this way? Simply put it is the innovation factor. I have in general seen very little (note "very little" not "none" before a flame war sparks) in the way of innovation from the free open source community. This is not to say that there are not quality products out there, but is more a statement of the drive to "catch up" and play "me too" against commercial vendors. What will drive innovation if commercial vendors pull out of the market?

Another concern I have is that there is so much fragmentation in the free open source world. Certainly one can argue that the great thing is that you can customize your packages however you like, but what tends to get glossed over is once you have customized things how do you smoothly upgrade to the next version? Is there enough interoperability being focused on between packages? The community as is has made great strides improving this, but I still think they have a ways to go when it comes to seamless interop that you get from the Windows / Office / etc suites where you can drag and drop from anywhere to anywhere. The additional problem with having such a diverse community is that in the standards argument there will always be groups that simply will not accept Direction A and will fork to Direction B which leads to more fragmentation and pain for users that pick Direction A which is eventually declared a failure and abandoned for Direction B.

Published Feb 09 2006, 08:05 AM by Eric Wise

Comments

Jeffrey Palermo said:

Innovation does come from everywhere. Microsoft is now trying to play catch up with testing tools like their new unit-testing framework. Microsoft is playing catch-up in browser innovation (not market share). Microsoft is playing catch-up in source control.

Also, the question of "is open source killing the ide/tools market": The question is a fallacy. Open source is a part of the market, not opposed to it. Open source offerings help make the market stronger. They are just another competitor. Market forces will make the best tool win, whether it's open source or for profit.

The market is constantly changing, and people who refuse to change will be left behind. Customers always have the freedom of where to spend money, so companies cannot feel that they _deserve_ to keep their customers. They must continue to serve them.

# February 9, 2006 12:35 PM

Eber Irigoyen said:

so you wouln't be interested in signing the "save Borland Programming Tools" petition...
http://www.petitiononline.com/savebpt/petition.html

# February 9, 2006 1:47 PM

Sam said:

Lack of innovation?

Apt/Yum/Ports?

Ruby/Rails, Python/Django?

The fastest database around for >90% of use cases?

The ease of management SSH, SwitchTower, Rake, etc brings to the party?

Eclipse's speed advantage over VS.NET, plus integrated plugin download/installation support with a documented interface and a wealth of extensions?

If you think OSS lacks innovation, I might suggest you're just not looking. Does Windows & MSOffice have significant advantages in many areas? Definitely!

And as an Open Office user at home, I don't know where you're getting your ideas. I think there's truly very _very_ VERY few people who couldn't live with the features OO has and _need_ MSOffice.

# February 9, 2006 2:34 PM

Peter's Gekko said:

This was found in a comment on Eric's post on open source and IDE's which refelected on Delphi for sale....

# February 10, 2006 4:48 AM

Peter's Gekko said:

This was found in a comment on Eric's post on open source and IDE's which refelected on Delphi for sale....

# February 10, 2006 5:12 AM

Eric Wise said:

From what I've seen in the OO world vs ms office excel totally crushes the OO spreadsheet. I don't consider office for word processing at all since there are a ton of alternatives and the only thing useful about office is the acceptance of the .doc format.

Ruby/Rails, python etc aren't really all that innovative. They are built to handle specific programming situations in a better way than the alternatives, they're not a "paradigm shifting" event in my opinion.

Not sure what you mean about the fastest database around? I know that mysql just recently got the ability to do some very basic things (subqueries for example).

Once again, I'm not bashing these products because they are good at what they do, as far as paradigm shifting innovation though none of these really fit the bill. Office/Exchange from lotus notes was a huge leap, these things are all sort of "me too" in my opinion.

# February 10, 2006 10:44 AM

Sam said:

Rails not innovative? Very smart people, some even on this site, have had trouble figuring out where to put validation code. That's just one example. I think maybe you should experience building a website with Rails before knocking it as a "me too".

MySQL is pretty much the fastest database around. I can't speak of the subquery support since I've only started toying with it in version 5, but it's fast, it's free. That's innovative I think.

Firefox is pretty innovative I think. You'd be right in pointing out that they weren't the first in a lot of areas, but they were the first Windows platform browser to get a lot of things right.

How about Eclipse and it's plugin support? Or it's refactoring support? Have you ever tried it?

Different strokes for different folks though I suppose. I have a hard time understanding what's supposed to set Outlook (which I use every day at work) apart, when something like Eclipse or Rails doesn't get the nod. Outlook certainly wasn't the first to do pretty much anything... but it was the first on Windows to get a lot of it right. That's innovation to me.

Or how about this for innovation: I can goto pretty much any *NIX, SSH into another machine, and administer it.

I can install MySQL on pretty much any platform I want, and access it from another platform with native drivers. Now that's the sort of innovation I'd like to see from Microsoft, and if they announced that sort of support, I for one certainly would consider that innovative. Atleast with respect to the history of SQL Server.

# February 10, 2006 5:09 PM

Eric Wise said:

Outlook/Office get the nod because of the interoperability between the various utilities. You can drag and drop a spreadsheet segment into an email etc. That was the first package that ever allowed you to do such things afaik.

Once again, MySQL is certainly catching up in functionality, but for a long time there was a huge tradeoff in functionality between mysql and the big boys. Postgre would be a better contender if you were going this route.

Eclipse is great, I never said anything negative about it in my post.

UNIX is a great example of a paradigm shift, it revolutionized things. Apple revolutionized the GUI. Windows and Linux builds are the me-too of this generation. They are both taking what they feel is the best from Apple and Unix and extending them.

As for ruby, I've toyed with it a bit, here's what I don't like about it:

1. Performance seems slow
2. Documentation just isn't there yet
3. I hate the syntax
4. When I played with it late last year, the XML support was shitty, no really, it was shitty (REXML).

# February 11, 2006 11:16 AM

Sam said:

OLE was around for some time before Outlook AFAIK.

You may not like Ruby or Rails (which you side-stepped), but that has nothing to do with innovation. I think Closures, Continuations, DSLs, "Extension Methods" are all pretty innovative.

... and you're welcome to your opinion of Ruby, but if you'd like to learn more I'd be happy to help.

As far as MySQL:

You know, I used to argue with the MySQL fanboi's too. They obviously just didn't "get it". Today I realize that those people were writing working applications while I was getting all frothy over different features in SQL Server that I, and better than 99% of websites around I'd bet, never used.

Still, old habits die hard I suppose... I prefer PostgreSQL even if it is about half the speed. ;-)

# February 12, 2006 4:03 AM

Eric Wise said:

That's really the thing about MySQL, it's free, but free isn't innovation, that's just a business model. At the root of things you have a database engine that does basic database goo very well, but little else in the way of advanced functionality.

This is where SQL Server, Oracle, etc are separating themselves from the pack. If you want things like reporting services, OLAP Cubes, etc then their solution is superior. If you're just running a few applications in a void without much in the way of replication etc then MySQL is an excellent choice. I just wouldn't call it "innovative".

The thing I've noticed about Languages, be it Ruby, C#, LINQ, or whatever is that is seems regardless of how far we have come as developers in tools and languages, we as an industry are still writing expensive, buggy code. The innovations in my mind in the development arena have been in the processes like AGILE, Test First Development, etc that help improve the quality of the code. Of course this thread was fun just because innovation is a perception thing so everyone will have varied opinions.

And I will admit that my perceptions of Ruby are based on some very short term playing. That's part of the busy developer trap though, if I can't pick up something quickly, it falls to the wayside which is why in spite of a bunch of new things in PHP, Python, Ruby, etc the old mainstays of ASP and Java still hold the largest chunk of the market.

# February 12, 2006 3:07 PM

Sam said:

Excellent response.

I agree, today free isn't innovation. But a few years ago, you could pickup PHP & MySQL and create a business at no software cost. MySQL changed the economics of database driven dynamic websites. That's where I think the innovation comes in. Another major area is the performance. It's not that everything should be about speed, but having the _option_ to use a less featurefull engine that provided a lot more performance is pretty innovative in my opinion. You don't see that choice in MSSQL or PostgreSQL AFAIK. And while MySQL doesn't have many of the "Big Iron" features of these servers, it does have things like scalar types, regular expression support, etc. Things that are lacking in some of the commercial databases, and things that provide an almost instant gratification. (I hate the idea of scalar types personally, but they're there, and I'm not going to say every implementation using them is flawed).

I absolutely agree with your comments on methodologies, but I think languages, frameworks, etc can play a role. Is Rail's directory structure the right solution for _everyone_? Probably not. But it's great for the rest of us. It's integrated testing, and guidance on how to build an application result in fewer bugs. I strongly believe in that, since that's been my experience.

In c# for example, most of my bugs are in the compiler cues, not the logic. I find that less code is more understandable, and less prone to containing bugs. Ruby lets me write much less code.

BTW, I could swear that PHP has the largest installed base in dynamic websites. I could be wrong. (I don't know PHP myself.) I'm not sure it's relevant either way, it just occurred to me.

You do have a point though. I never seem to find the time to get through all the books and languages I'd like to either.

Fortunately, Ruby was by far the easiest language to learn on my list, but my c# background played a large role in that I think. I think I'll put up some "learning Ruby" tips on my blog. I think the "lack of documentation" complaint is inaccurate, but I can see how people get that perception. Sometimes it can be a bit daunting to know where to look when first starting out.

I'd also be very interested in hearing what you didn't like about REXML if you're interested in discussing it, but I'll leave that up to you.

# February 12, 2006 3:50 PM

Eric Wise said:

My biggest complaint about REXML was that the API differs from the API of just about any other language I've experienced. It also seemed to suffer in performance and wasn't as complete as other APIs I've used. There are better libraries out there but like php moving away from the standard libraries can cause inconveniences in distributions. Internal only it's not a big deal though.

I believe that PHP does have the largest installed base but that figure is "everyone" not really the enterprise. In your fortune 500s php is a small blip. I think it includes all the phpbb installs etc. PHP is great in that there is a large base of templates available so you can just copy your way to a working website. What I hate about it is the old school ASP spaghetti code and that if you do anything that isn't standard default php install you can have problems finding a host. Plus they tend to break backwards compatibility in each new version.

# February 13, 2006 8:56 AM

Kenny said:

Let's just hope that a proper software vendor buys Delphi and continues to develop it as a commercial product. I use Delphi every day and I'm am very uncomfortable with the idea of it being open source. Look what happened to Turbopower - they were one of the biggest component vendors for Delphi, decided to pack it in & made all their products open source. How many of those open source projects are still being actively developed?

# March 14, 2006 7:22 PM

Name (required)

Your URL (optional)

Comments (required)

Remember Me?