Re: Calm my fears please. . . .
When doing web development of the CRUD type the lack of type checking and the incessant whining that results from wrong types is one of the biggest productivity boosts that can be had. On Mon, Mar 9, 2015 at 7:13 AM, Alexander Burger a...@software-lab.de wrote: Hi Lawrence, monads. Does the Lisp world, specifically picoLisp, have a response to these insistent Besserwisser people? Sure: KISS -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: Calm my fears please. . . .
Hi Lawrence, monads. Does the Lisp world, specifically picoLisp, have a response to these insistent Besserwisser people? Sure: KISS -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Calm my fears please. . . .
. . because all the ML-derivative languages (Haskell, F#, Ocaml) say you absolutely HAVE to have a strong type system in order to program with the big boys and girls. Likewise, they talk reverently about functors and monads. Does the Lisp world, specifically picoLisp, have a response to these insistent Besserwisser people? LB
Re: Calm my fears please. . . .
Thank you Alex for this very well formulated and thought out answer!@Lawrence:If you really want to do the kindergarden bullshit-contest, you could try to argue that picolisp is very strong typed, with just 3 types: symbols, lists and numbers.But as symbol is used for nearly everything, it doesn't really make a strong case.Learn about picolisp typing: http://software-lab.de/doc/ref.html#dataListen to Alex, he got it completely right.In the end, only the results count, not the language and methods.It depends on the people and the specific situation and goals what the "right" thing is.- Original Message - From: Alex Gilding [mailto:alex.gild...@talktalk.net] To: picolisp@software-lab.de Sent: Sun, 8 Mar 2015 15:01:58 + Subject: On 8 March 2015 at 13:52, Lawrence Bottorff borg...@gmail.com wrote: Does the Lisp world, specifically picoLisp, have a response to these insistent "Besserwisser" people? I sincerely hope not. Please don't turn this into a tribal thing. This is the sort of reasoning that led to the CL community's terrifying reputation. Lisps (mostly) use dynamic typing because their designers believe dynamic typing is more appropriate. Dynamic typing is not, however, superior just because it's used by Lisps. There is no need to "defend" anything with words when you could be doing so by demonstrating superior engineering results. Nobody is kicking in Lisp's door and taking away our programming licences because of the dynamic types (if someone is acting like that - maybe you have an annoying coworker IRL? - then they're not debating, they're being immature. Don't lower the discourse by taking it to their level). Just use the best tool for the job after carefully considering what exactly it is you need done. All paradigms and type systems have their valid place You wouldn't use picoLisp to write e.g. CompCert (you could, but it'd have to reinvent static types in order to do it; what a waste of energy). There is no need for the hammers to declare war on the screwdrivers. If you really need a counterargument favouring dynamic typing, consider that there's absolutely nothing about dynamic types that's specific to Lisp anyway: much bigger, more popular languages like Python, Ruby, JS etc. have the same attribute and are used to build massive, reliable systems all the time. They've already won this argument for you for all practical purposes.
Re: Calm my fears please. . . .
On 8 March 2015 at 13:52, Lawrence Bottorff borg...@gmail.com wrote: Does the Lisp world, specifically picoLisp, have a response to these insistent Besserwisser people? I sincerely hope not. Please don't turn this into a tribal thing. This is the sort of reasoning that led to the CL community's terrifying reputation. Lisps (mostly) use dynamic typing because their designers believe dynamic typing is more appropriate. Dynamic typing is not, however, superior just because it's used by Lisps. There is no need to defend anything with words when you could be doing so by demonstrating superior engineering results. Nobody is kicking in Lisp's door and taking away our programming licences because of the dynamic types (if someone is acting like that - maybe you have an annoying coworker IRL? - then they're not debating, they're being immature. Don't lower the discourse by taking it to their level). Just use the best tool for the job after carefully considering what exactly it is you need done. All paradigms and type systems have their valid place. You wouldn't use picoLisp to write e.g. CompCert (you could, but it'd have to reinvent static types in order to do it; what a waste of energy). There is no need for the hammers to declare war on the screwdrivers. If you really need a counterargument favouring dynamic typing, consider that there's absolutely nothing about dynamic types that's specific to Lisp anyway: much bigger, more popular languages like Python, Ruby, JS etc. have the same attribute and are used to build massive, reliable systems all the time. They've already won this argument for you for all practical purposes.
