Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.forth > #29767 > unrolled thread
| Started by | Mark Wills <markwills1970@gmail.com> |
|---|---|
| First post | 2014-04-22 03:02 -0700 |
| Last post | 2014-05-04 13:46 -0400 |
| Articles | 20 on this page of 123 — 32 participants |
Back to article view | Back to comp.lang.forth
One for Elizabeth... Mark Wills <markwills1970@gmail.com> - 2014-04-22 03:02 -0700
Re: One for Elizabeth... "Elizabeth D. Rather" <erather@forth.com> - 2014-04-22 09:04 -1000
Re: One for Elizabeth... Jecel <jecel@merlintec.com> - 2014-04-23 15:37 -0700
Re: One for Elizabeth... Bernd Paysan <bernd.paysan@gmx.de> - 2014-04-24 01:20 +0200
Re: One for Elizabeth... anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2014-04-24 17:24 +0000
Re: One for Elizabeth... "Rod Pemberton" <dont_use_email@xnothavet.cqm> - 2014-04-25 00:22 -0400
Re: One for Elizabeth... Mark Wills <markwills1970@gmail.com> - 2014-04-25 00:53 -0700
Re: One for Elizabeth... Howerd <howerdo@yahoo.co.uk> - 2014-04-25 05:16 -0700
Re: One for Elizabeth... anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2014-04-25 10:19 +0000
Re: One for Elizabeth... Andrew Haley <andrew29@littlepinkcloud.invalid> - 2014-04-25 09:40 -0500
Re: One for Elizabeth... anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2014-04-25 15:01 +0000
Re: One for Elizabeth... Andrew Haley <andrew29@littlepinkcloud.invalid> - 2014-04-25 10:57 -0500
Re: One for Elizabeth... anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2014-04-25 16:58 +0000
Re: One for Elizabeth... Paul E Bennett <Paul_E.Bennett@topmail.co.uk> - 2014-04-26 07:52 +0100
Re: One for Elizabeth... "Elizabeth D. Rather" <erather@forth.com> - 2014-04-25 09:01 -1000
Re: One for Elizabeth... Mark Wills <markwills1970@gmail.com> - 2014-05-06 01:17 -0700
Re: One for Elizabeth... Mark Wills <markwills1970@gmail.com> - 2014-05-06 01:16 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-04-23 22:12 -0700
Re: One for Elizabeth... Shark8 <OneWingedShark@gmail.com> - 2014-04-24 11:02 -0600
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-04-24 19:21 -0700
Re: One for Elizabeth... Shark8 <OneWingedShark@gmail.com> - 2014-04-24 21:52 -0600
Re: One for Elizabeth... "Rod Pemberton" <dont_use_email@xnothavet.cqm> - 2014-04-25 00:27 -0400
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-04-24 20:24 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-04-24 19:40 -0700
Re: One for Elizabeth... Shark8 <OneWingedShark@gmail.com> - 2014-04-24 21:57 -0600
Re: One for Elizabeth... "Bill Leary" <Bill_Leary@msn.com> - 2014-04-25 21:04 -0400
Re: One for Elizabeth... Elizabeth D Rather <erather@forth.com> - 2014-04-25 15:57 -1000
Re: One for Elizabeth... Steve Graham <jsgrahamus@yahoo.com> - 2014-04-26 11:49 -0700
Re: One for Elizabeth... "Elizabeth D. Rather" <erather@forth.com> - 2014-04-26 10:30 -1000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-04-27 17:55 -0700
Re: One for Elizabeth... stephenXXX@mpeforth.com (Stephen Pelc) - 2014-04-28 11:33 +0000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-04-28 20:53 -0700
Re: One for Elizabeth... stephenXXX@mpeforth.com (Stephen Pelc) - 2014-04-29 07:46 +0000
Re: One for Elizabeth... "Elizabeth D. Rather" <erather@forth.com> - 2014-04-28 21:57 -1000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-01 20:59 -0700
Re: One for Elizabeth... stephenXXX@mpeforth.com (Stephen Pelc) - 2014-05-02 07:46 +0000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-02 20:31 -0700
Re: One for Elizabeth... stephenXXX@mpeforth.com (Stephen Pelc) - 2014-05-03 08:26 +0000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-03 19:33 -0700
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-04 03:35 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-04 11:06 -0700
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-05 01:11 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-05 20:45 -0700
Re: One for Elizabeth... Rob Sciuk <rob@controlq.com> - 2014-05-06 12:53 -0400
Re: One for Elizabeth... Assad Ebrahim <assad.ebrahim@alum.swarthmore.edu> - 2014-05-07 09:59 +0100
Re: One for Elizabeth... rickman <gnuarm@gmail.com> - 2014-05-07 12:12 -0400
Re: One for Elizabeth... Coos Haak <chforth@hccnet.nl> - 2014-05-08 02:05 +0200
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-10 23:01 -0700
Re: One for Elizabeth... Paul Rubin <no.email@nospam.invalid> - 2014-05-05 20:58 -0700
Re: One for Elizabeth... AKK <akk@nospam.org> - 2014-05-06 07:16 +0200
Re: One for Elizabeth... "Elizabeth D. Rather" <erather@forth.com> - 2014-05-05 20:52 -1000
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-06 01:17 -0700
Re: One for Elizabeth... "Alex McDonald" <blog@rivadpm.com> - 2014-05-06 11:13 +0100
Re: One for Elizabeth... "Rod Pemberton" <dont_use_email@xnothavet.cqm> - 2014-05-06 07:12 -0400
Re: One for Elizabeth... Alex McDonald <blog@rivadpm.com> - 2014-05-06 10:57 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-06 17:06 -0700
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-06 07:52 -0700
Re: One for Elizabeth... Melzzzzz <mel@zzzzz.com> - 2014-05-06 17:11 +0200
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-06 08:57 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-07 00:38 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-07 22:56 -0700
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-08 04:10 -0700
Re: One for Elizabeth... thomas.worthington@london.ac.uk - 2014-05-08 04:42 -0700
Re: One for Elizabeth... rickman <gnuarm@gmail.com> - 2014-05-08 11:17 -0400
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-07 00:31 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-07 00:23 -0700
Re: One for Elizabeth... Paul Rubin <no.email@nospam.invalid> - 2014-05-07 02:29 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-07 13:39 -0700
Re: One for Elizabeth... Paul Rubin <no.email@nospam.invalid> - 2014-05-08 01:26 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-08 04:11 -0700
Re: One for Elizabeth... Paul Rubin <no.email@nospam.invalid> - 2014-05-08 08:34 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-08 11:04 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-08 17:28 -0700
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-09 02:55 -0700
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-09 10:42 -0700
Re: One for Elizabeth... Paul Rubin <no.email@nospam.invalid> - 2014-05-09 19:59 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-09 20:25 -0700
Re: One for Elizabeth... Laurent Laborde <kerdezixe@gmail.com> - 2014-05-11 05:24 -0700
Re: One for Elizabeth... Julian Fondren <ayrnieu@gmail.com> - 2014-05-12 22:08 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-12 22:11 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-13 22:37 -0700
Re: One for Elizabeth... "Rod Pemberton" <dont_use_email@xnothavet.cqm> - 2014-05-14 04:09 -0400
Re: One for Elizabeth... Stefan Mauerhofer <smauerhofer@androsoft.ch> - 2014-05-14 04:44 -0700
Re: One for Elizabeth... Mark Wills <markwills1970@gmail.com> - 2014-05-14 05:05 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-14 20:53 -0700
Re: One for Elizabeth... rickman <gnuarm@gmail.com> - 2014-05-15 01:08 -0400
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-18 22:23 -0700
Re: One for Elizabeth... Whammo <sidciavic@gmail.com> - 2014-05-19 09:48 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-21 19:50 -0700
Re: One for Elizabeth... "Rod Pemberton" <dont_use_email@xnothavet.cqm> - 2014-05-22 00:46 -0400
Re: One for Elizabeth... "Rod Pemberton" <dont_use_email@xnothavet.cqm> - 2014-05-19 22:37 -0400
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-21 23:58 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-21 19:24 -0700
Re: One for Elizabeth... albert@spenarnc.xs4all.nl (Albert van der Horst) - 2014-05-22 11:05 +0000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-22 23:19 -0700
Re: One for Elizabeth... albert@spenarnc.xs4all.nl (Albert van der Horst) - 2014-05-23 11:34 +0000
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-26 11:44 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-27 14:20 +0100
Re: One for Elizabeth... rickman <gnuarm@gmail.com> - 2014-05-27 10:30 -0400
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-27 19:58 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-28 07:45 +0100
Re: One for Elizabeth... Andrew Haley <andrew29@littlepinkcloud.invalid> - 2014-05-28 03:03 -0500
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-28 16:36 -0700
Re: One for Elizabeth... rickman <gnuarm@gmail.com> - 2014-05-28 23:03 -0400
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-29 11:01 -0700
Re: One for Elizabeth... anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2014-05-30 16:44 +0000
Re: One for Elizabeth... "Elizabeth D. Rather" <erather@forth.com> - 2014-05-29 08:25 -1000
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-21 23:32 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-26 11:15 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-27 19:42 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-27 20:23 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-28 09:14 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-28 16:25 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-29 20:47 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-29 20:28 -0700
Re: One for Elizabeth... Gerry Jackson <gerry@jackson9000.fsnet.co.uk> - 2014-05-30 22:24 +0100
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-30 21:49 -0700
Re: One for Elizabeth... Bernd Paysan <bernd.paysan@gmx.de> - 2014-05-29 23:40 +0200
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-29 19:47 -0700
Re: One for Elizabeth... Paul E Bennett <Paul_E.Bennett@topmail.co.uk> - 2014-05-08 14:36 +0100
Re: One for Elizabeth... visualforth@rocketmail.com - 2014-05-08 10:46 -0700
Re: One for Elizabeth... hughaguilar96@yahoo.com - 2014-05-07 22:24 -0700
Re: One for Elizabeth... Rob Sciuk <rob@controlq.com> - 2014-05-04 13:46 -0400
Page 1 of 7 [1] 2 3 4 5 6 7 Next page →
| From | Mark Wills <markwills1970@gmail.com> |
|---|---|
| Date | 2014-04-22 03:02 -0700 |
| Subject | One for Elizabeth... |
| Message-ID | <5c56c350-73d1-4a9d-9c39-509d8c27870b@googlegroups.com> |
At the bottom of Chuck Moore's Forth - The Early Years page (http://www.colorforth.com/HOPL.html) is the following interesting line: "The Forth story has the making of a morality play: Persistant young programmer struggles against indifference to discover Truth and save his suffering comrades. It gets better: Watch Forth. Inc go head to head with IBM over a French banking system." Just wandering if Elizabeth (or anyone else in the know) could shed any light on Forth Inc. and French banking system" story? Mark
[toc] | [next] | [standalone]
| From | "Elizabeth D. Rather" <erather@forth.com> |
|---|---|
| Date | 2014-04-22 09:04 -1000 |
| Message-ID | <ZvKdnQIbWYSgIMvOnZ2dnUVZ_rmdnZ2d@supernews.com> |
| In reply to | #29767 |
On 4/22/14 12:02 AM, Mark Wills wrote: > At the bottom of Chuck Moore's Forth - The Early Years page (http://www.colorforth.com/HOPL.html) is the following interesting line: > > "The Forth story has the making of a morality play: Persistant young programmer struggles against indifference to discover Truth and save his suffering comrades. It gets better: Watch Forth. Inc go head to head with IBM over a French banking system." > > Just wandering if Elizabeth (or anyone else in the know) could shed any light on Forth Inc. and French banking system" story? Happy to. It's a good story. In about 1979 we received a phone call from a Frenchman named Pierre Moreton. It was a Thursday. He said he was working with a French bank, SNVB, regarding the use of Forth on both IBM Series 1 minicomputers and banking terminals made by the Swedish firm Datasaab, and needed Chuck to be in Paris on Monday to do a presentation. However, Chuck was in the middle of a project, so on Friday it was decided that I would go, leaving Saturday night and arriving Sunday evening. It's an 11-hr flight from LA to Paris. I arrived about 6. They met me at DeGaulle, and took me to the hotel, where I had an hour to shower before they took me to a two-star restaurant to explain the situation. All I remember of the evening, through a cloud of jet lag and wine, was the fact that I astounded them all by eating escargot (which Americans do NOT do, they said), and the profiteroles for desert were so good I can still remember them. The next day I made my presentation and was grilled for 8 hours by several banking executives, 2 engineers from Datasaab, and 3 from IBM. Now, in those days, it was well known that *no one* bid successfully against IBM. If they wanted a project, they got it. At the time, FORTH, Inc. had a working polyFORTH on the Series 1 (multitasking, multiuser, extensive database support) as well as the 8080s used in the Datasaab terminals. I described it in detail. The lead IBM engineer said, "We have 300 programmers in Boca Raton, FL, working on software for the Series 1, and have a macro assembler working. How can you have all this?" I replied, "Because we don't try to do it with 300 programmers." At the end of the day, they said they needed me to repeat my presentation for more people from the bank, so I should stay a second day. I said I couldn't, because we had an important customer coming in on Wed. (thinking, that was a *real* customer, these people were a pipe dream). They offered to fly me home on the Concorde, with a schedule that would get me home by early afternoon Wed. I accepted. The Concorde left Paris at 12:30 pm, and arrived at JFK at 8:30 am. after 3 hrs of great food, wine, cognac. I reeled off the plane and onto the LA flight, and slept the whole way home. I somehow got through the meeting. The bank sent 2 executives and 2 IBM engineers to visit several of our Series 1 installations and talk with our customers, finally coming to LA to talk business. Everything we said they would see, they saw, including one installation with over 100 users on a S1. However, IBM flatly refused to maintain the S1 *hardware* if our software was on it, so the bank just hired us to program the Datasaab terminals (never bid against IBM). They sent us terminals, we programmed them, and Chuck and I spent a week in Linkoping, Sweden, at Datasaab and 2 weeks in France installing the system and training engineers. The head of the IBM sales team that had been involved was demoted and sent to a small town in Eastern France. Subsequently, Datasaab decided that they didn't want to use proprietary software, and some Swedish programmers attempted to program the Datasaab devices using a FIG variant. This was unsuccessful (I don't know any of the details), and Forth was dropped. Cheers, Elizabeth -- ================================================== Elizabeth D. Rather (US & Canada) 800-55-FORTH FORTH Inc. +1 310.999.6784 5959 West Century Blvd. Suite 700 Los Angeles, CA 90045 http://www.forth.com "Forth-based products and Services for real-time applications since 1973." ==================================================
[toc] | [prev] | [next] | [standalone]
| From | Jecel <jecel@merlintec.com> |
|---|---|
| Date | 2014-04-23 15:37 -0700 |
| Message-ID | <bef1c062-85ea-4cdd-ae62-a9306dfb1a1c@googlegroups.com> |
| In reply to | #29781 |
Elizabeth, great story! Thanks for telling it. > [...] The lead IBM engineer said, "We > have 300 programmers in Boca Raton, FL, working on software for the > Series 1, and have a macro assembler working. How can you have all > this?" I replied, "Because we don't try to do it with 300 programmers." They had heard something similar from Seymour Cray in the 1960s about the CDC 6600. It obviously hadn't sunk in, yet. -- Jecel
[toc] | [prev] | [next] | [standalone]
| From | Bernd Paysan <bernd.paysan@gmx.de> |
|---|---|
| Date | 2014-04-24 01:20 +0200 |
| Message-ID | <lj9hss$vj9$1@online.de> |
| In reply to | #29798 |
Jecel wrote: > Elizabeth, > > great story! Thanks for telling it. > >> [...] The lead IBM engineer said, "We >> have 300 programmers in Boca Raton, FL, working on software for the >> Series 1, and have a macro assembler working. How can you have all >> this?" I replied, "Because we don't try to do it with 300 programmers." > > They had heard something similar from Seymour Cray in the 1960s about the > CDC 6600. It obviously hadn't sunk in, yet. Somebody called Fred Brooks even wrote a book about that sort of miscalculation... and this guy even was employed by IBM ;-). Early march, I was meeting with other people who are creating a new internet (like my net2o, but bigger - for me, this is to exchange ideas, because I deliberately don't like the idea of "bigger"), and some sponsors, and one question from the sponsors was how long it would take if they had a billion euros. Well, with a billion euros, you can hire 10k people for a year, or 1000 people for 10 years. And then, whatever they do, will take a long time, because 1000 people create so much overhead that it takes very long. In this case, it's even worse: the more people you have, the less secure it will be. We know that you maybe can get down to one bug per 5000 LoCs. So if you have half a million LoCs (like GNUnet has), you'll finally have 100 bugs - if it is all well-debugged and written in a clean and straight- forward way (which GNUnet isn't). And even not very "severe" bugs can break security. -- Bernd Paysan "If you want it done right, you have to do it yourself" http://bernd-paysan.de/
[toc] | [prev] | [next] | [standalone]
| From | anton@mips.complang.tuwien.ac.at (Anton Ertl) |
|---|---|
| Date | 2014-04-24 17:24 +0000 |
| Message-ID | <2014Apr24.192401@mips.complang.tuwien.ac.at> |
| In reply to | #29799 |
Bernd Paysan <bernd.paysan@gmx.de> writes:
>In this case, it's even worse: the more people you have, the less secure it
>will be.
Not necessarily: Have one guy write the code, and let the other 999 do
code audits, demonstrate exploits etc. Of course there is the problem
that the feedback from the 999 overwhelms the one guy (including false
positives), but I guess with appropriate management practices this can
be handled.
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2013: http://www.euroforth.org/ef13/
[toc] | [prev] | [next] | [standalone]
| From | "Rod Pemberton" <dont_use_email@xnothavet.cqm> |
|---|---|
| Date | 2014-04-25 00:22 -0400 |
| Message-ID | <op.xeuugkzn6zenlw@localhost> |
| In reply to | #29808 |
On Thu, 24 Apr 2014 13:24:01 -0400, Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote: > Bernd Paysan <bernd.paysan@gmx.de> writes: >> In this case, it's even worse: the more people you have, the less >> secure it will be. > > Not necessarily: Have one guy write the code, and let the other 999 do > code audits, demonstrate exploits etc. Of course there is the problem > that the feedback from the 999 overwhelms the one guy (including false > positives), but I guess with appropriate management practices this can > be handled. > I don't understand that claim at all. Generally, companies pay the best and brightest they can get who program to program. So, code auditors clearly aren't the best and brightest. Otherwise, they'd be programming. Does having 999 chimpanzees check the work of Einstein prove anything? I.e., the things missing from this relationship are: intelligence, experience, skills, and the correct order and balance between them. Rod Pemberton
[toc] | [prev] | [next] | [standalone]
| From | Mark Wills <markwills1970@gmail.com> |
|---|---|
| Date | 2014-04-25 00:53 -0700 |
| Message-ID | <29c90e2c-3a8a-433f-843f-53dd50fa1549@googlegroups.com> |
| In reply to | #29816 |
On Friday, 25 April 2014 05:22:58 UTC+1, Rod Pemberton wrote: > I don't understand that claim at all. Generally, companies pay the best > and brightest they can get who program to program. So, code auditors > clearly aren't the best and brightest. Otherwise, they'd be programming. > Does having 999 chimpanzees check the work of Einstein prove anything? > I.e., the things missing from this relationship are: intelligence, > experience, skills, and the correct order and balance between them. I take your point, but it really can equally apply the other way around: You can have code monkeys banging out code all day (graduate/young engineers etc) and the real experienced guys are doing the code reviews and pulling the younger guys up when they make dumb mistakes - sort of passing on their knowledge (i.e. the code reviewers will be ex-programmers). In our industry (oil and gas) we have skilled professionals writing the code and the code and accompanying documentation (written to IEC61508 standards) is actually audited by professional, external audit teams. The auditors will typically have decades of experience and are very expensive. That's what you have to do when you want to write say SIL2 (probability of failure on demand < 10^-6) compliant software.
[toc] | [prev] | [next] | [standalone]
| From | Howerd <howerdo@yahoo.co.uk> |
|---|---|
| Date | 2014-04-25 05:16 -0700 |
| Message-ID | <d6ecddd3-248a-4c81-ab77-b0d9a0d7e32c@googlegroups.com> |
| In reply to | #29816 |
Hi Rod, > Generally, companies pay the best > and brightest they can get who program to program In my experience, companies are created, and must operate, to make a profit. People working within such a company tend to want an easy, comfortable life. Decisions are made according to these two contradictory pressures, and IMHO very rarely have anything to do with brightness, ability to program or intelligence. Many people here on CLF seem to me to express a similar view, that companies do the strangest things, and I find in reassuring to here these stories - I feel that I am not alone in failing to see the Emperor's new clothes... > Does having 999 chimpanzees check the work of Einstein prove anything? I agree that having 999 chimpanzees test a program is a pointless exercise, but this would not necessarily stop it happening in a large company ;-) Come to think of it, isn't this the norm (replacing "chimpanzees" by "junior programmers" )? After, all junior programmers are cheaper... Best regards, Howerd On Friday, 25 April 2014 05:22:58 UTC+1, Rod Pemberton wrote: > On Thu, 24 Apr 2014 13:24:01 -0400, Anton Ertl > > <a....@mips.complang.tuwien.ac.at> wrote: > > > Bernd Paysan <b.....p.....@gmx.de> writes: > > > > >> In this case, it's even worse: the more people you have, the less > > >> secure it will be. > > > > > > Not necessarily: Have one guy write the code, and let the other 999 do > > > code audits, demonstrate exploits etc. Of course there is the problem > > > that the feedback from the 999 overwhelms the one guy (including false > > > positives), but I guess with appropriate management practices this can > > > be handled. > > > > > > > I don't understand that claim at all. Generally, companies pay the best > > and brightest they can get who program to program. So, code auditors > > clearly aren't the best and brightest. Otherwise, they'd be programming. > > Does having 999 chimpanzees check the work of Einstein prove anything? > > I.e., the things missing from this relationship are: intelligence, > > experience, skills, and the correct order and balance between them. > > > > > > Rod Pemberton
[toc] | [prev] | [next] | [standalone]
| From | anton@mips.complang.tuwien.ac.at (Anton Ertl) |
|---|---|
| Date | 2014-04-25 10:19 +0000 |
| Message-ID | <2014Apr25.121912@mips.complang.tuwien.ac.at> |
| In reply to | #29816 |
"Rod Pemberton" <dont_use_email@xnothavet.cqm> writes:
>On Thu, 24 Apr 2014 13:24:01 -0400, Anton Ertl
><anton@mips.complang.tuwien.ac.at> wrote:
>> Bernd Paysan <bernd.paysan@gmx.de> writes:
>
>>> In this case, it's even worse: the more people you have, the less
>>> secure it will be.
>>
>> Not necessarily: Have one guy write the code, and let the other 999 do
>> code audits, demonstrate exploits etc. Of course there is the problem
>> that the feedback from the 999 overwhelms the one guy (including false
>> positives), but I guess with appropriate management practices this can
>> be handled.
>>
>
>I don't understand that claim at all. Generally, companies pay the best
>and brightest they can get who program to program. So, code auditors
>clearly aren't the best and brightest. Otherwise, they'd be programming.
>Does having 999 chimpanzees check the work of Einstein prove anything?
Actually, throwing random input at programs has proved surprisingly
effective at unveiling bugs.
Anyway, a company that hires 999 chimpanzees for programming will
likely put chimpanzees in charge of developing, auditing, and
mananging the whole thing. But I did not have such a company in mind.
If the 999 people are not chimpanzees, but competent programmers (even
if they are not geniuses), they will be able to look for and find
bugs.
The 1:999 setup was just an extreme example. It's probably better if
they have several competing efforts at producing the code, some with
single programmers, some with small teams, a significant number of
them auditing the code, others testing, etc.
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2013: http://www.euroforth.org/ef13/
[toc] | [prev] | [next] | [standalone]
| From | Andrew Haley <andrew29@littlepinkcloud.invalid> |
|---|---|
| Date | 2014-04-25 09:40 -0500 |
| Message-ID | <kKSdnW0bV9hu7sfOnZ2dnUVZ_q-dnZ2d@supernews.com> |
| In reply to | #29824 |
Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote: > "Rod Pemberton" <dont_use_email@xnothavet.cqm> writes: >>On Thu, 24 Apr 2014 13:24:01 -0400, Anton Ertl >><anton@mips.complang.tuwien.ac.at> wrote: >>> Bernd Paysan <bernd.paysan@gmx.de> writes: >> >>>> In this case, it's even worse: the more people you have, the less >>>> secure it will be. >>> >>> Not necessarily: Have one guy write the code, and let the other 999 do >>> code audits, demonstrate exploits etc. Of course there is the problem >>> that the feedback from the 999 overwhelms the one guy (including false >>> positives), but I guess with appropriate management practices this can >>> be handled. >>> >> >>I don't understand that claim at all. Generally, companies pay the best >>and brightest they can get who program to program. So, code auditors >>clearly aren't the best and brightest. Otherwise, they'd be programming. >>Does having 999 chimpanzees check the work of Einstein prove anything? > > Actually, throwing random input at programs has proved surprisingly > effective at unveiling bugs. > > Anyway, a company that hires 999 chimpanzees for programming will > likely put chimpanzees in charge of developing, auditing, and > mananging the whole thing. But I did not have such a company in mind. > > If the 999 people are not chimpanzees, but competent programmers (even > if they are not geniuses), they will be able to look for and find > bugs. > > The 1:999 setup was just an extreme example. It's probably better if > they have several competing efforts at producing the code, some with > single programmers, some with small teams, a significant number of > them auditing the code, others testing, etc. I am perhaps more of a believer in co-operation that competitions of this kind. I have never participated in such a competition, and I'm sure I would find it very demotivating. Your comments remind me of the famous "super-programmer project" paper by J. D. Aron, abstracted here: Harlan Mills, the first of IBM's chief programmers, was assigned a thirty-man-year project, to be finished in six months - by himself! For a variety of reasons, including the fact that he assembled a group of helpers for specific aspects of the project, it took six man-years to complete the job. Nevertheless, the project seemed successful enough such that IBM employed the approach again in the famous New York Times project [see Papers 7 and 11] and formalized the Chief Programmer Team concept for use throughout the entire company. Even this first experimental project contained many of the basic concepts of the formalized Chief Programmer Team. For example, Harlan Mills had not one, but two, people who normally would be called "program librarians'" One was a secretary, the other a trainee programmer. He also used a "language lawyer" to help deal with the intricacies of the PL/I programming language. Perhaps the most important part of Aron's paper is the last paragraph, in which he deals with "the question of why we don't use this technique in our many other projects." Aron's explanation is crucial, because it describes precisely why most other EDP organizations have abandoned the chief programmer concept after a period of initial fascination: " . . . because we cannot depend upon the existence of a super-programmer in any given project and, because we must transfer our people around in order to apply the right number of resources at the right time to a contract, we have found that it is important to build our systems without dependence upon any particularly strong individual." http://dl.acm.org/citation.cfm?id=1241519 Andrew.
[toc] | [prev] | [next] | [standalone]
| From | anton@mips.complang.tuwien.ac.at (Anton Ertl) |
|---|---|
| Date | 2014-04-25 15:01 +0000 |
| Message-ID | <2014Apr25.170107@mips.complang.tuwien.ac.at> |
| In reply to | #29825 |
Andrew Haley <andrew29@littlepinkcloud.invalid> writes:
>Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote:
>> The 1:999 setup was just an extreme example. It's probably better if
>> they have several competing efforts at producing the code, some with
>> single programmers, some with small teams, a significant number of
>> them auditing the code, others testing, etc.
>
>I am perhaps more of a believer in co-operation that competitions of
>this kind. I have never participated in such a competition, and I'm
>sure I would find it very demotivating.
The problem as posed is: If you want to waste a lot of man-years on a
project that should be finished in a short time, how do you go about
it? One idea is to exploit search parallelism by letting different
teams work at it. They will choose different approaches and/or take
different design decisions; most of them will finish faster than one
project that tries to employ all 1000 people in one project (unless
there is embarassingly parallel work in the project), and some will
finish much faster.
We work that way in many areas, including programming: In science
research groups work non-exclusively on research topics that they
perceive relevant, different companies work on products independently
and in parallel, different programming projects work on solving the
same or similar problems (just today I heard a talk where the
presenter said that there is a new web framework every week).
>Your comments remind me of the famous "super-programmer project" paper
>by J. D. Aron, abstracted here:
Sure, that's the idea Bernd is promoting. It certainly seems quite
appropriate if you want to finish quickly.
> Perhaps the most important part of Aron's paper is the last
> paragraph, in which he deals with "the question of why we don't use
> this technique in our many other projects." Aron's explanation is
> crucial, because it describes precisely why most other EDP
> organizations have abandoned the chief programmer concept after a
> period of initial fascination:
>
> " . . . because we cannot depend upon the existence of a
> super-programmer in any given project and, because we must transfer
> our people around in order to apply the right number of resources at
> the right time to a contract, we have found that it is important to
> build our systems without dependence upon any particularly strong
> individual."
>
>http://dl.acm.org/citation.cfm?id=1241519
Which resonates with many things we have read here about why Forth is
not more popular.
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2013: http://www.euroforth.org/ef13/
[toc] | [prev] | [next] | [standalone]
| From | Andrew Haley <andrew29@littlepinkcloud.invalid> |
|---|---|
| Date | 2014-04-25 10:57 -0500 |
| Message-ID | <wNGdnTG82_5HGMfOnZ2dnUVZ8vydnZ2d@supernews.com> |
| In reply to | #29826 |
Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote: > Andrew Haley <andrew29@littlepinkcloud.invalid> writes: >>Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote: >>> The 1:999 setup was just an extreme example. It's probably better if >>> they have several competing efforts at producing the code, some with >>> single programmers, some with small teams, a significant number of >>> them auditing the code, others testing, etc. >> >>I am perhaps more of a believer in co-operation that competitions of >>this kind. I have never participated in such a competition, and I'm >>sure I would find it very demotivating. > > The problem as posed is: If you want to waste a lot of man-years on a > project that should be finished in a short time, how do you go about > it? One idea is to exploit search parallelism by letting different > teams work at it. They will choose different approaches and/or take > different design decisions; most of them will finish faster than one > project that tries to employ all 1000 people in one project (unless > there is embarassingly parallel work in the project), and some will > finish much faster. Oh sure, that's true, but I would be worried that good people would say "This is stupid" and either not try very hard or leave to do something else. Motivation is key in all tasks. > We work that way in many areas, including programming: In science > research groups work non-exclusively on research topics that they > perceive relevant, different companies work on products > independently and in parallel, different programming projects work > on solving the same or similar problems (just today I heard a talk > where the presenter said that there is a new web framework every > week). Indeed. Having said that, I suspect that people co-operate in science more than thay compete, but I accept the idea that it is very important for people to try to replicate or falsify other people's results. Andrew.
[toc] | [prev] | [next] | [standalone]
| From | anton@mips.complang.tuwien.ac.at (Anton Ertl) |
|---|---|
| Date | 2014-04-25 16:58 +0000 |
| Message-ID | <2014Apr25.185813@mips.complang.tuwien.ac.at> |
| In reply to | #29828 |
Andrew Haley <andrew29@littlepinkcloud.invalid> writes:
>Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote:
>> The problem as posed is: If you want to waste a lot of man-years on a
>> project that should be finished in a short time, how do you go about
>> it? One idea is to exploit search parallelism by letting different
>> teams work at it. They will choose different approaches and/or take
>> different design decisions; most of them will finish faster than one
>> project that tries to employ all 1000 people in one project (unless
>> there is embarassingly parallel work in the project), and some will
>> finish much faster.
>
>Oh sure, that's true, but I would be worried that good people would
>say "This is stupid" and either not try very hard or leave to do
>something else. Motivation is key in all tasks.
Competition is also a great motivator. Whether this approach is
presented more as a competition or more as uncoordinated parallel
development of different approaches for the same problem is a matter
of emphasis. That's up for management to choose.
One question there is: do you want the various groups to work openly
(so they can inspire each other), or in secret (so they do not fall
prey to the same mistake which they got from a particularly convincing
presentation)? If the latter, make it more competetive, with bonuses
for the winners etc.
>> We work that way in many areas, including programming: In science
>> research groups work non-exclusively on research topics that they
>> perceive relevant, different companies work on products
>> independently and in parallel, different programming projects work
>> on solving the same or similar problems (just today I heard a talk
>> where the presenter said that there is a new web framework every
>> week).
>
>Indeed. Having said that, I suspect that people co-operate in science
>more than thay compete, but I accept the idea that it is very
>important for people to try to replicate or falsify other people's
>results.
Yes, but that's not what I was referring to; doing that is much more
cooperative than when you work on your own research topic. And while
there is certainly a lot of cooperation in science, we all learn from
each other and pick up other peoples ideas and integrate them into our
work, there is also the issue that different research groups often
work on pretty much the same topics in parallel. There is also a more
explicitly competetive element, particularly in CS: At a prestigious
conference only maybe 30 papers out of maybe 150 submissions is
accepted; research funding institutions accept only the best research
proposals for funding.
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2013: http://www.euroforth.org/ef13/
[toc] | [prev] | [next] | [standalone]
| From | Paul E Bennett <Paul_E.Bennett@topmail.co.uk> |
|---|---|
| Date | 2014-04-26 07:52 +0100 |
| Message-ID | <bs13diFl593U1@mid.individual.net> |
| In reply to | #29831 |
Anton Ertl wrote: [%X] > Competition is also a great motivator. Whether this approach is > presented more as a competition or more as uncoordinated parallel > development of different approaches for the same problem is a matter > of emphasis. That's up for management to choose. > > One question there is: do you want the various groups to work openly > (so they can inspire each other), or in secret (so they do not fall > prey to the same mistake which they got from a particularly convincing > presentation)? If the latter, make it more competetive, with bonuses > for the winners etc. In some Safety Critical Systems where you need the agreement of several channels the "competitive secret effort" approach from a goal based specification can yield sufficient slightly different software that you have a valid voting from the results of diverse logic. Of course, you may have select n results out of n+x efforts in a sane review process. Added to that, the voting system then becomes the most critical part of the design as you can only ever have one of those. [%X] -- ******************************************************************** Paul E. Bennett IEng MIET.....<email://Paul_E.Bennett@topmail.co.uk> Forth based HIDECS Consultancy.............<http://www.hidecs.co.uk> Mob: +44 (0)7811-639972 Tel: +44 (0)1235-510979 Going Forth Safely ..... EBA. www.electric-boat-association.org.uk.. ********************************************************************
[toc] | [prev] | [next] | [standalone]
| From | "Elizabeth D. Rather" <erather@forth.com> |
|---|---|
| Date | 2014-04-25 09:01 -1000 |
| Message-ID | <ZZ-dnZaaTPeQLMfOnZ2dnUVZ_oWdnZ2d@supernews.com> |
| In reply to | #29828 |
On 4/25/14 5:57 AM, Andrew Haley wrote: > Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote: >> Andrew Haley <andrew29@littlepinkcloud.invalid> writes: >>> Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote: >>>> The 1:999 setup was just an extreme example. It's probably better if >>>> they have several competing efforts at producing the code, some with >>>> single programmers, some with small teams, a significant number of >>>> them auditing the code, others testing, etc. >>> >>> I am perhaps more of a believer in co-operation that competitions of >>> this kind. I have never participated in such a competition, and I'm >>> sure I would find it very demotivating. So would any of the excellent programmers I have worked with. >> The problem as posed is: If you want to waste a lot of man-years on a >> project that should be finished in a short time, how do you go about >> it? Don't. Just, don't. If you want the project finished in the shortest possible time, get the best programmer you know, be really clear about the requirements and the objective, and be prepared to provide any assistance he/she requests. >> One idea is to exploit search parallelism by letting different >> teams work at it. They will choose different approaches and/or take >> different design decisions; most of them will finish faster than one >> project that tries to employ all 1000 people in one project (unless >> there is embarassingly parallel work in the project), and some will >> finish much faster. 1,000 people on a project can't do anything. If it's a really big project, it can be factored into a lot of small projects, but it has to be done carefully and the interfaces well-defined. If the boundaries between the sub-projects aren't obvious, maybe it shouldn't be divided that way, etc. > Oh sure, that's true, but I would be worried that good people would > say "This is stupid" and either not try very hard or leave to do > something else. Motivation is key in all tasks. Absolutely. As far as the testing goes, there are people who excel at it, and delight in coming up with diabolical challenges to throw at the code to see if it will break. It's a skill set and motivation all its own. Look for those people and cherish them! Cheers, Elizabeth -- ================================================== Elizabeth D. Rather (US & Canada) 800-55-FORTH FORTH Inc. +1 310.999.6784 5959 West Century Blvd. Suite 700 Los Angeles, CA 90045 http://www.forth.com "Forth-based products and Services for real-time applications since 1973." ==================================================
[toc] | [prev] | [next] | [standalone]
| From | Mark Wills <markwills1970@gmail.com> |
|---|---|
| Date | 2014-05-06 01:17 -0700 |
| Message-ID | <60238fc3-76fa-4b38-8050-d4417d5b7029@googlegroups.com> |
| In reply to | #29799 |
On Thursday, 24 April 2014 00:20:59 UTC+1, Bernd Paysan wrote: > Somebody called Fred Brooks even wrote a book about that sort of > miscalculation... and this guy even was employed by IBM ;-). Touché! Just catching up with this thread...
[toc] | [prev] | [next] | [standalone]
| From | Mark Wills <markwills1970@gmail.com> |
|---|---|
| Date | 2014-05-06 01:16 -0700 |
| Message-ID | <a24b33ac-f41d-4f6b-92ba-da809e161116@googlegroups.com> |
| In reply to | #29798 |
On Wednesday, 23 April 2014 23:37:43 UTC+1, Jecel wrote: > They had heard something similar from Seymour Cray in the 1960s about the CDC > 6600. It obviously hadn't sunk in, yet. They also had Fred Brooks saying the same thing from within IBM. Perhaps they should have read his book ;-)
[toc] | [prev] | [next] | [standalone]
| From | hughaguilar96@yahoo.com |
|---|---|
| Date | 2014-04-23 22:12 -0700 |
| Message-ID | <962890b8-d773-4d03-a5aa-f14c016eaf4f@googlegroups.com> |
| In reply to | #29781 |
On Tuesday, April 22, 2014 12:04:29 PM UTC-7, Elizabeth D. Rather wrote:
> In about 1979 we received a phone call from a Frenchman named Pierre
> Moreton. It was a Thursday. He said he was working with a French bank,
> SNVB, regarding the use of Forth on both IBM Series 1 minicomputers and
> banking terminals made by the Swedish firm Datasaab, and needed Chuck to
> be in Paris on Monday to do a presentation. However, Chuck was in the
> middle of a project, so on Friday it was decided that I would go,
> leaving Saturday night and arriving Sunday evening.
"Chuck was in the middle of a project" --- means that he was busy writing software, which is what programmers normally do with their time.
> It's an 11-hr flight from LA to Paris. I arrived about 6. They met me at
> DeGaulle, and took me to the hotel, where I had an hour to shower before
> they took me to a two-star restaurant to explain the situation. All I
> remember of the evening, through a cloud of jet lag and wine, was the
> fact that I astounded them all by eating escargot (which Americans do
> NOT do, they said), and the profiteroles for desert were so good I can
> still remember them.
>
> At the end of the day, they said they needed me to repeat my
> presentation for more people from the bank, so I should stay a second
> day. I said I couldn't, because we had an important customer coming in
> on Wed. (thinking, that was a *real* customer, these people were a pipe
> dream). They offered to fly me home on the Concorde, with a schedule
> that would get me home by early afternoon Wed. I accepted. The Concorde
> left Paris at 12:30 pm, and arrived at JFK at 8:30 am. after 3 hrs of
> great food, wine, cognac. I reeled off the plane and onto the LA flight,
> and slept the whole way home. I somehow got through the meeting.
When I was working as a C/C++ programmer, I one time got to fly to LA to install our latest release of the software and train the workers there on using the new features. My boss told me that he couldn't afford to give programmers bonuses, but since the customer was idiot enough to pay for somebody to fly out there, he would let me do it as my bonus. Normally I would have just typed out a couple of pages of documentation on how to install the software and use it, and emailed that to the customer, but apparently they are illiterate in LA and needed somebody to come out and tell them verbally. Anyway, the trip was super-fun! I had a company credit-card in my back pocket and could go to restaurants that I wouldn't ordinarily be able to afford. I went to one Mexican restaurant and had lobster tacos --- not very authentic as there aren't any lobsters in Mexico --- but really delicious. I didn't drink any cognac (I don't even know what that is) --- I just drank a Dos Equis (if it is good enough for the "most interesting man in the world," then it is good enough for me). I seriously considered going to a brothel, but I decided not to stretch my luck that much --- lobster tacos on the credit-card might be okay, but the accounting lady would be unlikely to accept a charge at a brothel. If it had been Paris rather than LA I would have gone for it though --- American whores are all skanks ("skank" is a French word for a woman who doesn't use the bidet after sex, so the next guy gets "sloppy seconds").
When I got back to Phoenix I used my story about going to LA to impress girls, and that worked fairly well. In my experience, women don't understand the concept of "pride of accomplishment" --- for women, it is all about results, such as a fast car or a condo in Hawaii. Women are very focused on paying for things with other people's money, so when I described having the "company credit-card in my back pocket," they were very impressed.
> Subsequently, Datasaab decided that they didn't want to use proprietary
> software, and some Swedish programmers attempted to program the Datasaab
> devices using a FIG variant. This was unsuccessful (I don't know any of
> the details), and Forth was dropped.
The reason why people don't want to use proprietary closed-source software is because they run into bugs in the software. Because they don't have the source-code, they can't fix the bugs themselves. When they complain about the bugs, a sales-person tells them that they need to pay for an upgrade to the software.
This was my experience with Forth Inc.. SwiftForth version-2 had severe bugs. When I complained, Elizabeth Rather hit me with a sales-pitch for buying version-3 for another $300. This was my first encounter with Elizabeth Rather, and this is what first turned me off from her. It was just sleazy --- this is most likely why Datasaab dropped Forth Inc. too.
I knew some programmers in Denver who had written a compiler for the 6502 a long time ago. Recently (somewhere around 2000) they got an email from some astronomers in Turkey who were still using a 6502 motion-control board to aim their telescope, and their software was written for this compiler. The Turks complained that they had upgraded their board from 32K to 48K of RAM (the prom was 16K), and they had discovered that the language couldn't make use of the extra 16K of RAM. They wanted the bug fixed. The programmers had sold this software decades ago and had forgotten about it. But they were honorable men, so they dug out the source-code for their compiler and fixed it so that it could make use of the extra memory, and they shipped the fixed software to Turkey (they had to find a floppy disk drive in storage so they could make a disk). This is how real programmers are --- they support their software. It is sleazy salespeople who demand money for every "upgrade," which is really just a bug-fix.
[toc] | [prev] | [next] | [standalone]
| From | Shark8 <OneWingedShark@gmail.com> |
|---|---|
| Date | 2014-04-24 11:02 -0600 |
| Message-ID | <Ysb6v.149355$Ng5.81289@fx17.iad> |
| In reply to | #29804 |
On 23-Apr-14 23:12, hughaguilar96@yahoo.com wrote: > It is sleazy salespeople who demand money for every "upgrade," which is really just a bug-fix. Ah, but it works the other way, too: Customers can, and do, try to sneak new features into "bug-fixes"... especially if the bug-fix rate is less than the rate for new development.
[toc] | [prev] | [next] | [standalone]
| From | hughaguilar96@yahoo.com |
|---|---|
| Date | 2014-04-24 19:21 -0700 |
| Message-ID | <579b0f22-1eb4-461e-8dfb-6c35e2e2dca0@googlegroups.com> |
| In reply to | #29806 |
On Thursday, April 24, 2014 10:02:47 AM UTC-7, Shark8 wrote: > On 23-Apr-14 23:12, hughaguilar96@yahoo.com wrote: > > > It is sleazy salespeople who demand money for every "upgrade," which is really just a bug-fix. > > Ah, but it works the other way, too: > Customers can, and do, try to sneak new features into "bug-fixes"... > especially if the bug-fix rate is less than the rate for new development. I never heard of a "bug-fix rate" --- I think any respectable programmer is going to fix bugs for free as part of the warranty, if he has been paid for the software that he wrote. The distinction between a bug-fix and a feature can be vague though. In the case of those guys with the 6502 compiler, if the astronomers had wanted to access memory beyond the basic 64K limit using bank-switching, that would be a new feature. The astronomers were pretty justified in expecting to access the full 64K though, so unless the documentation explicitly said that there was a 32K limit, the programmers had to admit that their 32K limit was a bug and fix the problem for free. SwiftForth version-2 had severe bugs (the most egregious being that (LOCAL) crashed the system when it was used). That clunker of a compiler should never have been sold for money, and after it was sold, Forth Inc. really should have provided their customers with a fixed version for free. Forth Inc. just keeps scamming people decade after decade, and they are ruining Forth's reputation in the world. I worked as an IBM370 assembly-language programmer for almost 2 years. We had CSRs (customer service reps) who would get specs from the customers and pass those on to the programmer. Quite often, the specs I got did not make any sense, and were not even self-consistent. So I would be my own CSR, doing both jobs and only getting paid for one, while the CSR would flit around like a butterfly wringing her hands. We did the data-processing for mass mailings, and most of our customers were scam artists --- they would send out junk mail soliciting donations for starving children in Africa, or to cure cancer, or whatever, and they were obviously just pocketing all the money themselves. They would routinely try to scam me with "feature inflation." With minor changes in the specs, I would just do it, but they would try to take advantage of this to push through as much as they could. I found that it was very important to always have a specs sheet, and to require them to sign off on any changes to it. A favorite tactic of theirs was to verbally get me to make changes from the specs, and then to complain that my software wasn't to spec and that they deserved a discount because of this. Getting them to sign the spec sheet was like trying to catch a squirrel with your bare hands, but I wouldn't write a line of code until I had the signature, so they had to relent. They would tell me that they golf with my boss every weekend and could get me fired with a casual word. They were lying --- my boss hadn't golfed in years due to a bad back, but he still had golfing memorabilia in his office which is how they knew he was interested in the game. I told my boss about all of this hassle, and he said: "Better you than me." He said that he was well aware of what kind of people the customers were, and that he paid me to deal with them so that he didn't have to. He said not to worry about getting into trouble though, as he stood behind me all the way. He eventually retired though, and that was the end of that.
[toc] | [prev] | [next] | [standalone]
Page 1 of 7 [1] 2 3 4 5 6 7 Next page →
Back to top | Article view | comp.lang.forth
csiph-web