Re: Competition

On 08/26/2011 11:47 PM, Scrap Happy wrote:
>> True. My supervisor mainly cares about the results. He doesn't care much
>> what software I use to analyze my data or create my plots as long as it
>> works (I use a Perl script to do the analysis and pipe the results to
>> Gnuplot).
>
> That sounds unusually tolerant. Are you living in Academia, perchance?

Yup, academia. Astrophysics to be exact. I am a grad student interested 
in theoretical work, and especially computer simulations. Right now I'm 
studying rocky planets like Earth and how likely they are to stay in a 
nice stable orbit when the gas giants go bezerk.

Earlier I worked for a small company making web applications. Again, I 
had a lot of freedom as long as the apps ran fine in all browsers. So I 
haven't had the experience of truly having to integrate tools the way 
you do.


> If I draft a schematic, I have to ensure that the PCB layout tool
> can *read* that schematic. The output (photoplot) from the PCB tool
> has to be comprehensible to the PCB fabricator. The part list from
> the schematic tool has to be compatible with the MRP system the
> client has in place. etc.
>
> It is rare that something will exist in isolation, for me (even
> paper documents need to be transported to print shops, posted on
> web pages, etc. -- this affects the fonts used, document format,
> etc.)


I see. Yeah, that's worlds apart from the things I do. With your work 
everything has to fit perfectly or else the work is useless.



> Yes. For example, any "pro bono" work that I do (typically, 500-1000
> hours annually) -- where *I* have exclusive control of how things are
> done -- tends to gravitate towards FOSS. It places no demands on those
> that will eventually have to maintain my work later (by contrast, if I
> had to take a commercial tool and ensure its continued availability at
> some future date, then I -- or someone else -- has to ante up the money
> to purchase that license...)
>
> And, it (hopefully) entices that future maintainer to look into
> something (FOSS) that he/she might not otherwise have done (since
> it is a lot easier to maintain an existing system with its existing
> toolchain than to risk porting the whole thing to a different
> toolchain -- esp when that existing toolchain is "free"!)


Very interesting.


-- 
I'm not overweight, I'm undertall.