Subject: Re: Latest benchmarks on Tcl vs. Python? - DN [1]
David Oppenheimer <davidopp@megsinet.net> - 02 Sep 1999 - comp.lang.tcl
--------------FAF5B2340C2A80A6241C7914 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I am not sure how valid benchmark testing is, let's try to remember the recent fiasco where Microsoft placed a highly optimized M$ server against a poorly optimized Linux server (remember those Mindcrap tests?). It may be more useful to ask what tasks the various scripting languages are better at. No single language can honestly claim to be the Holy Grail (notice the Monty Python tie-in?) due to the inherent limitations in each language. Python enthusiasts will openly admit that there are sections of code that they will write in C/C++ due to the runtime advantages. Likewise, device drivers and some other kinds of software cannot be programmed in an interpreted scripting language (even if speed of code execution is not an issue). Instead of trying to create divisions, I think that cross pollination and convergent evolution is more productive than one upmanship. Instead of saying "Your language sucks, I could do that in 3 fewer lines and with a run-time savings of 20%," I believe people would be better off by saying, "Your (insert language name) code could run faster if you could implement a (insert language name)-like method of doing that." Let's keep in mind here that there is much to be gained by cooperation and an open exchange of ideas. Of course, for those who have fond memories of grade school and long for name calling, there are always flame wars. Here are a couple of links: From '95 comparing Python & Tcl: http://www.cwi.nl/~sjoerd/PythonVsTcl.html From '98 from someone who switched from Tcl to Python: http://www.uk.research.att.com/~fms/ipc7/tr-1998-9.html From the most recent O'Reilly Open Source convention Hear the leading members of the open source community talk about the state of open source in streaming Real Audio: http://www.technetcast.com/tnc_program.html?program_id=38 Now that the flame war with the PERL community seems to have subsided, let's hope that the Python & Tcl communities can be a little more mature. It's OK to say I prefer one language over another. To say someone else's language sucks is just plain wrong. Let's keep in mind that with the many GUI designer's available that Pythons in charge decided to incorporate Tkinter (which uses Tcl) as its defacto standard GUI maker. (think cooperation & cross pollination people!!!) David O. "Ron A. Zajac" wrote: > I just did an Altavista search, but didn't come across any recent > performance benchmarks comparing tcl and python. I found an old tcl7.6 > comparison, and I'm sure tcl performs better than that now, with the > type conversion caching. And I'm curious. Anyone got a URL? --------------FAF5B2340C2A80A6241C7914 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html> <tt>I am not sure how valid benchmark testing is, let's try to remember the recent fiasco where Microsoft placed a highly optimized M$ server against a poorly optimized Linux server (remember those Mindcrap tests?). It may be more useful to ask what tasks the various scripting languages are better at. No single language can honestly claim to be the Holy Grail (notice the Monty Python tie-in?) due to the inherent limitations in each language. Python enthusiasts will openly admit that there are sections of code that they will write in C/C++ due to the runtime advantages. Likewise, device drivers and some other kinds of software cannot be programmed in an interpreted scripting language (even if speed of code execution is not an issue).</tt><tt></tt> <p><tt>Instead of trying to create divisions, I think that cross pollination and convergent evolution is more productive than one upmanship. Instead of saying "Your language sucks, I could do that in 3 fewer lines and with a run-time savings of 20%," I believe people would be better off by saying, "Your (insert language name) code could run faster if you could implement a (insert language name)-like method of doing that."</tt><tt></tt> <p><tt>Let's keep in mind here that there is much to be gained by cooperation and an open exchange of ideas. Of course, for those who have fond memories of grade school and long for name calling, there are always flame wars. </tt> <p><tt>Here are a couple of links:</tt><tt></tt> <p><tt> From '95 comparing Python & Tcl:</tt> <br><tt><a href="http://www.cwi.nl/~sjoerd/PythonVsTcl.html">http://www.cwi.nl/~sjoerd/PythonVsTcl.html</a></tt>; <br><tt> From '98 from someone who switched from Tcl to Python:</tt> <br><tt><a href="http://www.uk.research.att.com/~fms/ipc7/tr-1998-9.html">http://www.uk.research.att.com/~fms/ipc7/tr-1998-9.html</a></tt>; <br><tt> From the most recent O'Reilly Open Source convention</tt> <br><tt>Hear the leading members of the open source community talk</tt> <br><tt>about the state of open source in streaming Real Audio:</tt> <br><a href="http://www.technetcast.com/tnc_program.html?program_id=38">http://www.technetcast.com/tnc_program.html?program_id=38</a><a href="http://www.technetcast.com/tnc_program.html?program_id=38"></a>; <p><tt>Now that the flame war with the PERL community seems to have subsided, let's hope that the Python & Tcl communities can be a little more mature. It's OK to say I prefer one language over another. To say someone else's language sucks is just plain wrong. Let's keep in mind that with the many GUI designer's available that Pythons in charge decided to incorporate Tkinter (which uses Tcl) as its defacto standard GUI maker. (think cooperation & cross pollination people!!!)</tt> <p>David O. <br> <p><tt>"Ron A. Zajac" wrote:</tt> <blockquote TYPE=CITE><tt>I just did an Altavista search, but didn't come across any recent</tt> <br><tt>performance benchmarks comparing tcl and python. I found an old tcl7.6</tt> <br><tt>comparison, and I'm sure tcl performs better than that now, with the</tt> <br><tt>type conversion caching. And I'm curious. Anyone got a URL?</tt></blockquote> </html> --------------FAF5B2340C2A80A6241C7914--
Last modified
1999-09-27
1999-09-27
(195.108.246.50)
Note: you are looking at
the snapshot of an old wiki
- much of this information
is likely to be very outdated