Re: [OpenSTA-devel] Open Source Load Testing

[Corey G. <cgo...@ea...>]

>I'd have a look at how the project works, how it gains momentum, what
>features it brings and how easy it is to submit patches. It's certainly
>easier to support one codebase than two that have once been very similar...

absolutely.. I maintain several open source projects and understand that forking usually only occurs out of necessity.  My point was really to let the OpenSTA devs and users know that there is another project with a lot of overlap with what OpenSTA does.  A lot of people like OpenSTA and some ideas have been tossed around regarding future development. so I don't wanna be the "hey everyone jump ship!" guy :)

Hopefully a good community will be fostered around WebLOAD.. however.. if that is not the case.. or Radview doesn't support the project properly.. the code is GPL'ed and is free to be forked.  Also, a fork doesn't have to occur for it be used in further OpenSTA development.  I'm not familiar with the WebLOAD architecture yet... but I noticed they have good documentation and I dove the source code for 20 mins or so (they have subversion repository up).  If the code is somewhat modular, parts can be used as libraries, shared between tools (think OpenSTA engine hooking into a WebLOAD monitoring module or whatever)... who knows where this could go.  There doesn't seem to be a huge number of developers interested in perf/load tools as has been seen in other Open Source tool communities (OpenSTA, JMeter, Grinder, etc).  possibly since this is starting from such a mature code base it could be different.. but who knows.  

I wrote a short blog post about this:  http://www.goldb.org/goldblog/2007/04/11/RadviewWebLOADGoesOpenSource.aspx

-Corey