From psmith@mozart.convex.com Wed Mar 18 21:33:22 1992 Received: from convex.convex.com by dkuug.dk via EUnet with SMTP (5.64+/8+bit/IDA-1.2.8) id AA20351; Wed, 18 Mar 92 21:33:22 +0100 Received: from mozart.convex.com by convex.convex.com (5.64/1.35) id AA09415; Wed, 18 Mar 92 14:31:53 -0600 Received: by mozart.convex.com (5.64/1.28) id AA26310; Wed, 18 Mar 92 14:31:51 -0600 Date: Wed, 18 Mar 92 14:31:51 -0600 From: psmith@mozart.convex.com (Presley Smith) Message-Id: <9203182031.AA26310@mozart.convex.com> To: maine@altair.dfrf.nasa.gov, sc22wg5@dkuug.dk Subject: Re: (SC22WG5.70) Re: Another tangent Cc: psmith@mozart.convex.com X-Charset: ASCII X-Char-Esc: 29 >I've stayed out of this thread so far because it seems to have too much the >flavor of a flame war and too little useful content. Temptation wins again, >though. I'll give in to the extent of one posting and then I'll try to >shut back up. Come on in. When Walt and I agree on some things, we are making progress. :-) >psmith@mozart.convex.com (Presley Smith) said: Presley> We really should try to get a picture of what the Fortran community Presley> looks like... Presley> I'll start the process. My Fortran base is made up of customers who's Presley> prime concern is PERFORMANCE!... >I represent a viewpoint almost diametrically opposed. I am, of >course, concerned with performance, but it is distinctly not my >highest priority. My highest priorities are divided more among areas >like code reliability and maintainability and programmer productivity. >These are why I view things like modules and several of the other new >f90 features as big improvements. >I have very little interest in code that runs very fast, but gets >wrong results. Believe me, I have seen many examples of such codes, >sometimes codes that have been in production use giving wrong results >for years. I view it as a very serious concern, not an academic debate. >I don't intend to argue what set of priorities is "best"; such >arguments tend to become pointless flame wars and this particular one >is not new. I only mean to suggest that there does exist a >significant group of users with priorities like mine. I am one and I >know others. I also know users with priorities more like those >Presley describes. I won't invent unsupportable numbers or >percentages. Great. I believe you are exactly right. There is a such a group that is interested in reliability, maintainability, and programmer productivity. What I'm saying is that if you asked 1000 Fortran programmers at random from multiple sites, discplines, etc. how they would rank performance, productivity, reliability and 100 other things... What would rank and the top and what would rank at the bottom. We have claims of "representing" the Fortran community. That's a diverse community that we don't have a good feel for what the overall priority of that community is. I've NEVER claimed to represent any part of the community other that the ones that I talk to in the business that we are in. The vast majority that I talk to would even change languages if that would improve their performance. That's all I'm saying... Presley> How big a group is this? It's public information that we have about Presley> 1000 computers with 500 customers. If we take 50 programmers per Presley> customer, which is probably low, then that equals about 25,000 Presley> Fortran users who's main concern is performance. >This is awfully speculative. We jump from a presumably accurate number >of systems through a speculation about the numbers of programmers, >an assumption that all of these are Fortran programmers, add a >blanket claim that all (or at least some unspecified large percentage) >of these have performance as their primary concern, and we get a number. I certainly did not mean to say that this was an accurate number. It's just an algorithm to try and measure something. I could get a more accurate idea by sending a questionaire to the customer base or asking for that information at our User Group meetings. My point is that we need to have some frame of reference as to what the community is and what the real concerns are. >I simply can't believe that all of Presley's customers are quite so >unanimous. I can't find such unanimity in my own branch, much less >such an impressive number as 25,000 people. I think I'm counted among >Presley's 25,000. NASA Ames has several Convex machines and our site >has been considering one. (Though I've recently been getting more >insistent with their sales reps that if I can't run f90 on the >machine, I'm not interested - trying to play my part in this "market" >that will decide the true fate of f90). If I was counted, perhaps you >should revise that figure to 24,999. We're right in sync with the vendor released Fortran 90 products to date. Every new version of our compiler has more Fortran 90 features added and the next one will also... And I'm fine with a number of 10,000... If you tell me you're a workstation vendor and there are 3 people running on each diskless workstation, then I have a problem believing that. We need to get some information into our process about what the Fortran community really looks like and what they want in a new standard. I form an opinion of my community based on talking to people who are a part of it during my business dealings. We need to find a way to expand that through multiple people and collect some real data for the desires for the next revision... We need to do a market survey!!! This is an analytical group of people... and yet we're all tied up in emotion without fact... We need to get back to our roots and analyze the Fortran market needs. >I am in the process of joining X3J3 and one of my primary reasons for >doing so is to help represent what I see as user concerns. I have >no allusions that any one person can represent the entire spectrum of >users. I would hate to see Fortran become too narrowly aimed at any >one niche. Yet balancing disparate concerns such as performance >versus robustness is not always easy. (And adding politics to the >brew doesn't make it any easier). I don't have simple answers and >I've probably blathered on for long enough. Great... we need more users on X3J3... Too many of us "nasty vendors" have been in the way of progress. :-) FYI. Presley