Fog Creek Software
Discussion Board




Commoditize Your Complements

Now that FogBUGZ runs with MySQL, I think it's time to take the next step - port the application so that it doesn't require NT/2000/XP. Specifically, I am suggesting porting the application to perl. By porting FogBUGZ to perl, you make the application platform agnostic. Your clients can choose to run the application using XP/IIS or Linux/Apache.

This isn't a "You must use Linux at all costs" rant - this is a "use the right tool for the job" rant. There is no real benefit in requireing NT/2000/XP and IIS for FogBUGZ. By making the application platform agnostic, it can be argued that you would expand your pool of potential clients.

I suppose the same argument could be made for porting FogBUGZ to java/jsp. I suggested perl because of my own experience with using the language on both Windows and Unix.

Let your clients pick the platform.

Wayne Earl
Monday, February 03, 2003

Better off waiting till they port to .NET and then just use Mono/Mono-Basic et al to make it all automagically work <g>

GiorgioG
Monday, February 03, 2003

I Don't think there are any plans of porting ASP.NET to linux. 

Vincent Marquez
Monday, February 03, 2003

There are plans to support ASP.NET in Mono, which will run on Linux.  See http://www.go-mono.com/asp-net

Jason Howat
Monday, February 03, 2003

Presumably, it would cost time and money to port our entire product to another programming language. So it's not as one sided as Wayne makes it out: it's a cost/benefit analysis.

Joel Spolsky
Monday, February 03, 2003

It's a shame that it would be too much of an ordeal to port it.  I'd buy it in a heartbeat.

Patrick Lioi
Monday, February 03, 2003

Isn't the plan that Mono will provide .NET support for *nix platforms (at least Linux and Mac OS X)?

That's my understanding and I hope Sun are going to do something about Java GUIs because at the point Mono is up and running Java loses its major competitive edge over .NET - platform independence - and the performance of Java GUIs will come under greater scrutiny.

Walter Rumsby
Monday, February 03, 2003

Of course, it would cost time and money to port the application - I never suggested it wouldn't. I just think that the benifits are greater than the cost.

There are two classes of bug tracking software in the Unix/Linux world: the Free Software route (bugzilla, et al) which are kludgy and can be difficult to set up and maintain. Then there is the very expensive commercial packages, which tend to target very large scale efforts (or at least, very large budgets). I think FogBUGZ is a reasonable middle ground between these two, and that there might be a market here for FogCreek.

Look at the rapid growth of Linux on the server, particularly in education, small business, and government. IMO, to port the application makes good business sence, but Joel certainly knows his market better than I do.

Wayne Earl
Monday, February 03, 2003

Wayne said..
---
There are two classes of bug tracking software in the Unix/Linux world: the Free Software route (bugzilla, et al) which are kludgy and can be difficult to set up and maintain. Then there is the very expensive commercial packages, which tend to target very large scale efforts (or at least, very large budgets). I think FogBUGZ is a reasonable middle ground between these two, and that there might be a market here for FogCreek.
---

I've been hitting up against that for the past few months.  We started using bugzilla, and I've already spent a couple days trying to get it to auto-change the bugs with any cvs checkins.  I've used CVS and coded perl for the web (cgi and mod_perl) for 3 years now, so its not that I don't know how these tools work.  Its just that I don't think bugzilla is very high quality / polished.  Too much custom code to get anything useful done.

Sure its got lots of features, but I don't want that as much as I want something easy to set up and that gets the job done.  FogBUGZ might fit that ...

The only feature that is missing in FogBUGZ (please prove me wrong! :) as far as I know is the ability to have different projects and groups, and keep them separate.  I want to have 2 people that can only see the Foo and Bar projects, a 3rd that can see the Baz project, and the group leader than can see all projects.  Everyone can only modify, see, and submit bugs in their own categories.  ACLs with groups for bug tracking, essentially.

We have an IIS server hanging out so thats no problem, and now that FB runs on MySQL, thats one more bird down, so the above feature is all I'm waiting on before I lobby my boss for a 10 person license (maybe more).  (note, I don't want to run many installs of FogBUGZ, one for each project).

Lastly, a script that would convert a bugzilla database to a FogBUGZ database, along with the above feature, would win you a *lot* of converts.  I'm as big a fan of open source as the next guy (I've worked on 2 OSS projects so far, and am about to start on a 3rd) so I like that aspect of bugzilla, but its not polished enough for me.  I don't want to spend 1 day a month futzing with it, and another 3 days fixing the UI.  Though if I could spend one day replacing it with a tool that fits our needs, and I don't have to mess with, I'd do it in a heartbeat.

I have a feeling that there are more than a few people that feel the same way as me.

Andrew Hurst
Tuesday, February 04, 2003

*  Recent Topics

*  Fog Creek Home