Mailing List Archive: 49091 messages
  • Home
  • Script library
  • AltME Archive
  • Mailing list
  • Articles Index
  • Site search
 

REBOL Benchmark test?

 [1/5] from: tim::johnsons-web::com at: 11-Aug-2001 10:02


Hello All: I hope to be going "commercial" with rebol this fall. That is, providing programming services with rebol. Where, I live rebol is virtually unknown. To convince IT managers that I might contract to, that rebol is viable, I know that they are going to be looking for testing procedures.... Are there benchmark tests available? I would find that helpful, because if I'm "allowed" to use rebol, then I might end up being able to bill for as many hours as I code :>) Comments and codes are welcome... Thanks Tim

 [2/5] from: dness:home at: 11-Aug-2001 13:38


Tim Johnson wrote:
> Hello All: > I hope to be going "commercial" with rebol this fall.
<<quoted lines omitted: 10>>
> Thanks > Tim
There's nothing wrong in being `armed' with benchmark results, but I'd be _very_ surprised if some notion of `efficiency' as measured by simple benchmarks of performance would be of _any_ substantial importance in making programming services purchase decisions. Indeed, focussing on such things is largely a sign of naiveté, IMO. In some `mission critical' applications, performance may be vitally important, but it strikes me as _highly_ unlikely that anyone would hand over a mission critical task to anyone without a long established reputation, so I think you are---at least for awhile---unlikely to be in the running for such jobs. Sensible managers are likely to spend most of their time worrying about how they can manage, understand and _support_ particular programming environments. I'd suggest you prepare yourself to answer along these dimensions rather than with a bunch of benchmark comparisons. In other words, think about answering questions like `What happens to the project if you get hit by a bus?' rather than `How does the timing of that inner-loop compare to execution in complied C code?'

 [3/5] from: dvydra:orion-it at: 11-Aug-2001 10:41


Specifically what kind of benchmark are you talking about? At 10:02 AM 8/11/2001 -0800, you wrote:

 [4/5] from: tim:johnsons-web at: 11-Aug-2001 11:19


On Sat, Aug 11, 2001 at 01:38:18PM -0400, David Ness wrote:
> There's nothing wrong in being `armed' with benchmark results, but I'd be > _very_ surprised if some notion of `efficiency' as measured by simple benchmarks > of performance would be of _any_ substantial importance in making programming > services purchase decisions. > > Indeed, focussing on such things is largely a sign of naiveté, IMO.
I agree with you 100%... However, naivete among IT managers CAN come with the territory, and I have to deal with it. ==>>Except for programming, all the work I have ever done has been out of doors. I never tried to change the weather! Since I have my own business, I can pick who I work for, and I am indeed picky, but sometimes I'll forgive a little naivete :>) Thank you for the advice, but I'm still looking for testing procedures, but no big deal, I can make them up myself - lest I be misunderstood, I've been programming for a living for 12 years now :>) Regards -- Tim

 [5/5] from: tim:johnsons-web at: 11-Aug-2001 11:40


Hi David.. On Sat, Aug 11, 2001 at 10:41:43AM -0700, David Vydra wrote:
> Specifically what kind of benchmark are you talking about?
timing, memory usage... Actually, as I think about it, just about everything I would be interested in has been covered in this list. I believe that the list is archived, but I've misplaced the link to it..... does anyone have it? I have also noted some references to differences in rebol's memory usage as per the OS. And you know, I'm a real rebol booster, but I have found that some ISP's are suspicious of rebol because it is new I've encountered a couple who wanted some sort of demonstration that rebol is stable, not a back door, etc. And I know of one large ISP in my area that uses windows 2000. They have so far refused to install rebol, because they go to http://www.rebol.com/platforms.html and see a binary for Windows 95/98/NT but not for '2000. awe... t'heck wid 'em? Where's that link to the list archive? I promise not to lose it this time. thnx Tim

Notes
  • Quoted lines have been omitted from some messages.
    View the message alone to see the lines that have been omitted