September 2019 - Pharo-users

Re: [Pharo-users] FFI beginner question

by Guillermo Polito

Just more detail into it: - the source code of the booklet, written in Pillar, resides in here: https://github.com/SquareBracketAssociates/Booklet-uFFI <https://github.com/SquareBracketAssociates/Booklet-uFFI> - Iâve been in the last weeks doing a pass on it (see branch version2 https://github.com/SquareBracketAssociates/Booklet-uFFI/tree/version2 <https://github.com/SquareBracketAssociates/Booklet-uFFI/tree/version2>) I started adding examples, and enhancing the explanations. I got particularly blocked when explaining marshalling where I saw several issues to address / features to add before continuing documenting I have some issues written down to fix in FFI soon: 1 - unify vocabulary: module and library in the API (see #ffiCall:module: #macModuleName #ffiLibraryName, #ffiLibraryâ¦) 2 - some of the names above are misleading (#ffiLibraryName returning a library object and not a name for example) 3 - extend literal object support to floats 4 - add a strict (but safe) mode where types for literals are mandatory (otherwise marshalling can go wrong :)) Iâll create issues for these in the next hours ^^. > El 23 sept 2019, a las 13:38, Richard O'Keefe <raoknz(a)gmail.com> escribiÃ³: > > how do I > connect C numeric > types introduced by the library to FFI? Hi Richard, what do you mean? Do you have an example? Whatâs the signature of your C function look like? Guille

Sept. 23, 2019

Re: [Pharo-users] Pharo-users Digest, Vol 77, Issue 67

by Richard O'Keefe

On Mon, 23 Sep 2019 at 16:49, Vince Refiti <vinref(a)gmail.com> wrote: > The SQLite3 API is very well documented, and the UDBC-SQLite3 project (https://github.com/astares/Pharo-UDBC) is a nice and clean binding to it. Look in UDBCSQLite3Library class. Thanks. It's nearly midnight here; I've downloaded it and shall look at it tomorrow.

Sept. 23, 2019

Re: [Pharo-users] FFI beginner question

by Richard O'Keefe

Thank you for that. I note that some sections have nothing in them. Now that I have read that, what is the next thing to read? In particular, how do I connect C numeric types introduced by the library to FFI? On Mon, 23 Sep 2019 at 16:45, TomaÅ¾ Turk <tomaz.turk(a)ef.uni-lj.si> wrote: > > Hi, > > The book draft is here: https://files.pharo.org/books-pdfs/booklet-uFFI/UFFIDRAFT.pdf. > > Best wishes, > Tomaz > > > > Any plans to draft a Pharo booklet on this subject? > > -Ted >

Sept. 23, 2019

Re: [Pharo-users] Defensive programming or not

by Richard O'Keefe

This is perhaps the biggest question I've faced working on my own Smalltalk, and it is connected to an annoyance. My answer is that when I am *using* library code, I want my mistakes to be detected at "interface" level. If, for example, I do ((1 to: 5) asOrderedCollection) at: 2.5 put: #fred; yourself I want the mistake to be detected and I want it to be reported as coming from OrderedCollection>>at:put: To my shock and dismay, Pharo just answered an OrderedCollection(1 #fred 3 4 5). VisualWorks, VisualAge, ST/X, Dolphin, and GNU ST all raise an exception if the index is not an integer. Can anyone explain to me why ^ index isNumber ifTrue: [ self at: index asInteger put: value] ifFalse: [ self errorNonIntegerIndex] ]. is thought to be an improvement over ^ self errorNonIntegerIndex ]. when it comes to Object>>[basic]at:[put:]? (The name #errorNonIntegerIndex suggests that other numbers were not originally allowed.) One thing I expect from an array-like object is that if i and j are different and x and y are different, then anArray at: i put: x; at: j put: y; at: i should answer x. But (Array new: 1) at: 1.1 put: #tin; at: 1.2 put: #pan; at: 1.1 is #pan, not #tin. To return to my initial point. Assuming a version of Smalltalk in which sequence indices *are* checked, I want (anOrderedCollection at: 2.5) to report the error as coming from *this* message, not some internal message sent to some internal object I have no normal access to. The annoyance is that this matters less than it should because while the exception *mechanism* is defined by the ANSI Smalltalk standard, almost no *exceptions* were defined, so there are in practical terms no standard exceptions to raise or handle. What I did for my own Smalltalk was (1) notice that a check (index between: 1 and: size) let Fractions, ScaledDecimals, FloatE, FloatD, FloatQ, DualNumbers, and QuadraticSurds slip through. (2) face-palm "how stupid of me! of COURSE that's not right" (3) introduce a new fast primitive (index integer_between: 1 and: size) where x integer_between: y and: z turns into (((x)|(y)|(z))&TAG) == 0 && (SignedWord)(y) <= (SignedWord)(x) && (SignedWord)(x) <= (SignedWord)(z) which I suppose should really be called something like #basicBetween:and:. (4) Discover that there were far more places where I should have been using integer_between:and: than I had originally expected. It's about a 50-50 split. My point here is that if it seems as though it might be too expensive to check, perhaps you have the wrong tools for checking. Or perhaps it is possible to weaken your precondition so that you don't need to report an errror. As an example of that, I ran into a problem in Squeak where collection1 removeAll: collection2 went crazy if collection1 and collection2 were the same. My library checks for that and makes it work. We sometimes say "Use the source, Luke!" But that's no good if Luke needs to know something that isn't *in* the source. So if a method has a precondition that is not a consequence of the class invariant and is not checked, it really needs to be commented. On Mon, 23 Sep 2019 at 20:19, Kasper Osterbye <kasper.osterbye(a)gmail.com> wrote: > > Cheers all, > > In trying to fix one of the issues in Pharo, I came to wonder about the prefered style in in dealing with preconditions to methods in Pharo. > > In the method Random>>nextInt: anInteger > > There is a check if anInteger is negative > There is no check that anInteger is anInteger > > When should I put a guard there, and throw an early warning, and when is it part of the (often implicit pre-condition)? > > Best, > > Kasper

Sept. 23, 2019

Re: [Pharo-users] Code of Conduct

by Offray

I'm not interpreting anything, particularly I'm not interpreting that part, as I just point to the whole FAQ without making a particular interpretation of snippets of it. Do your PR in the repo. I'm pretty sure that the bulk of your contributions to Pharo and the force of your arguments and specific suggestions will be considered by those who lead the project, as they have been in the case of others. El 22 de septiembre de 2019 10:14:26 PM GMT-05:00, Steve Quezadas <steveeq1(a)gmail.com> escribiÃ³: >> But the low rate at which marginalized people are recruited, and >> the high rate at which they leave the industry ><https://www.kaporcenter.org/tech-leavers/>, point to a larger >> cultural and systemic problem. > >Your interpreting this information with a SJW lens. Otherwise known as >"confirmation bias". Look at the low proportion of blacks and women >who >apply for CS majors in college. Are you going to say that colleges are >using discriminatory practices to keep blacks and women from taking CS >classes? Maybe the bulk of the low recruitment statistics is simply due >to >non-interest within that sub-culture. > >I believe this CoC is a way to wedge left-wing politics in a >non-political >maillist. I want it out. > >On Sun, Sep 22, 2019 at 7:37 PM Offray Vladimir Luna CÃ¡rdenas < >offray.luna(a)mutabit.com> wrote: > >> I agreed that the last decision should be on the ones who made the >bulk >> of the work. But I don't see relationship between a code of conduct >and >> not being able to talk about code or contributions quality. Just >looking >> at the FAQ of the original CoC that originated the whole think, I see >a >> lot of answers about the stuff being said on this thread (minorities, >> left wing progressive agenda, diminish of code quality because of it, >> mixing tech with non-tech stuff), so I will refer to it, because as I >> said, I think that the PR should be the place for the bulk of the >> discussion: >> >> https://www.contributor-covenant.org/faq >> >> The FAQ name goes pretty well, considering the amount of repeated >> arguments they deal with. I think that many of the FAQ apply for >other >> CoCs, despite of the possible different nature of CoC for the online >> community and the CoC for other face to face events. BTW, Thanks for >the >> links, both provide a better context for the emergence of the CoC in >the >> Erlang community. >> >> As said, I will try to see for specific contributions in the >> correspondent PR in the repo, and made some if I have a one. For the >> moment I'm trying to make my contributions on this thread, but is >taking >> a lot. >> >> Cheers, >> >> Offray >> >> On 22/09/19 7:40 p. m., Richard O'Keefe wrote: >> > This is not a question of left vs right. It's a question of >> > authoritarian vs libertarian. >> > And this is very relevant to the community. >> > It's also not a question of democracy vs central authority. >> > It's a question of vs ÏÎ±ÏÏÎ·ÏÎ¯Î± vs goodspeak. >> > And this is very relevant to the community also. >> > >> > Pharo is "owned" by the people who do the bulk of the work on it, >> > and they are kind enough to share it with us. That there is such a >> > thing as a *Pharo* community is the result of their work. >> > >> > That there is such a thing as a Pharo *community* depends on the >ability >> of >> > that community to communicate freely. This cuts BOTH ways. If >people >> are >> > scared off by incivility, that's bad. If people are driven away by >> incivility, >> > that's bad. But when you stop seeing rudeness as rudeness, which >may be >> > amended, and start seeing it as crimethink, you drive people away, >and >> that >> > is bad too. >> > >> > Let's consider a recent thread. I took the position that << and >putOn: >> were >> > confusing, unreliable, and unnecessary. The unreliability issue >has been >> > addressed in Pharo 8; had I not been able to speak I would never >have >> learned >> > that. Some people apparently think that it improves readability, >where >> I find >> > that << impairs my ability to understand. The fact that BOTH sides >were >> able >> > to speak freely means that we now know (a) that there is no >consensus for >> > removing them from the system and (b) if you want other people to >read >> your >> > code you might want to think twice before using them, and we are >all >> better off. >> > But if criticising someone's opinion were construed as harassment, >the >> thread >> > would have been shut down before I displayed my code with a >> generalisation >> > that is worth having if << is worth having at all. >> > >> > I probably should have mentioned the Erlang code of conduct >> > http://erlang.org/download/erlang_org_code_of_conduct.txt >> > It is pretty a-political, has graduated response, and potential for >> forgiveness. >> > >> > A code of conduct for *events* is another matter, which is why I >bring >> > Erlang up. >> > http://erlang.org/pipermail/erlang-questions/2015-March/083849.html >> > is eye-opening. (It's mainly about Ruby community issues.) >> > >> > On Mon, 23 Sep 2019 at 11:51, Offray Vladimir Luna CÃ¡rdenas >> > <offray.luna(a)mutabit.com> wrote: >> >> My point was that this community, as a the big majority of FLOSS >ones, >> is not a democracy and *not* having a democracy has shown its >benefits in >> human endeavors like science, technology, hackerspaces and so on. >> >> >> >> I'll keep the rest of the conversation with you on the source code >> repository and the PR. See you there. >> >> >> >> >> >> On 22/09/19 6:40 p. m., Steve Quezadas wrote: >> >> >> >> This isn't science, this is a community. We don't need a CoC, >there >> haven't been any problems on this list regarding nazis or whatever. >This is >> just a group of people trying to enforce their political ideologies >on >> everyone else. Let's just remove the CoC altogether and just replace >it >> with one line: "this maillist is about Pharo, anything else is >offtopic". >> >> >> >> If you want to debate on the merits of Islam vs Christianity/ >right vs >> left / thugs vs racists , you are free to hold your opinion on some >other >> sub, but it's offtopic here. >> >> >> >> On Sun, Sep 22, 2019 at 4:23 PM Offray Vladimir Luna CÃ¡rdenas < >> offray.luna(a)mutabit.com> wrote: >> >>> There is no data to support such supposed majority. But even so, >free, >> libre, open source communities are not democracies. Imagine the >quality of >> code or argumentation based on perceived majorities? If science would >be a >> democracy, the earth would be "still" flat. >> >>> >> >>> On 22/09/19 6:04 p. m., Steve Quezadas wrote: >> >>> >> >>> I would say that the majority don't seem to be in favor of it. >This >> should be a democracy. >> >>> >> >>> On Sun, Sep 22, 2019 at 1:53 PM Offray Vladimir Luna CÃ¡rdenas < >> offray.luna(a)mutabit.com> wrote: >> >>>> >> >>>> On 22/09/19 3:38 p. m., Steve Quezadas wrote: >> >>>>>> The discussion so far shows that CoC is not a distraction to >many >> >>>>> Actually, the discussion shows that the CoC is "a distraction >to >> many". >> >>>> Actually it shows that some people consider it a distraction, >others >> >>>> don't. I think that every body here is able to form its own >opinion on >> >>>> that and invest time and effort accordingly. >> >>>> >> >>>> Cheers, >> >>>> >> >>>> Offray >> >>>> >> >>>> >> >>>> >> > >> >> >> -- Enviado desde mi dispositivo Android con K-9 Mail. Por favor, disculpa mi brevedad.

Sept. 23, 2019

Re: [Pharo-users] Defensive programming or not

by Sven Van Caekenberghe

Hi Kasper, > On 23 Sep 2019, at 10:18, Kasper Osterbye <kasper.osterbye(a)gmail.com> wrote: > > Cheers all, > > In trying to fix one of the issues in Pharo, I came to wonder about the prefered style in in dealing with preconditions to methods in Pharo. > > In the method Random>>nextInt: anInteger > â¢ There is a check if anInteger is negative > â¢ There is no check that anInteger is anInteger > When should I put a guard there, and throw an early warning, and when is it part of the (often implicit pre-condition)? > > Best, > > Kasper This is very good question. In my opinion, in a dynamically typed language you should not explicitly enforce all types. It is OK to rely on MessageNotUnderstood exceptions, it is the ultimate protection. Of course, you could help your users/callers by throwing more useful exceptions, when that makes sense. There is also a cost with checking (too often). Furthermore, #isInteger or #isString are close testing the for a class, which is not considered good object design. In the case of Random>>nextInt: speed seems important, but a robust interface with type checking has some value too. Hard to decide. Sven

Sept. 23, 2019

Defensive programming or not

by Kasper Osterbye

Cheers all, In trying to fix one of the issues in Pharo, I came to wonder about the prefered style in in dealing with preconditions to methods in Pharo. In the method Random>>nextInt: anInteger - There is a check if anInteger is negative - There is no check that anInteger is anInteger When should I put a guard there, and throw an early warning, and when is it part of the (often implicit pre-condition)? Best, Kasper

Sept. 23, 2019

Re: [Pharo-users] Pharo-users Digest, Vol 77, Issue 67

by Vince Refiti

Re: FFI beginner question The SQLite3 API is very well documented, and the UDBC-SQLite3 project ( https://github.com/astares/Pharo-UDBC) is a nice and clean binding to it. Look in UDBCSQLite3Library class. Vince On Mon, Sep 23, 2019 at 1:15 PM <pharo-users-request(a)lists.pharo.org> wrote: > Send Pharo-users mailing list submissions to > pharo-users(a)lists.pharo.org > > To subscribe or unsubscribe via the World Wide Web, visit > > http://lists.pharo.org/mailman/listinfo/pharo-users_lists.pharo.org > or, via email, send a message with subject or body 'help' to > pharo-users-request(a)lists.pharo.org > > You can reach the person managing the list at > pharo-users-owner(a)lists.pharo.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pharo-users digest..." > > > Today's Topics: > > 1. FFI beginner question (Richard O'Keefe) > 2. Re: Code of Conduct (Richard O'Keefe) > 3. Re: FFI beginner question (Brainstorms) > 4. Re: Code of Conduct (Offray Vladimir Luna C?rdenas) > 5. Re: Code of Conduct (Steve Quezadas) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 23 Sep 2019 11:56:50 +1200 > From: "Richard O'Keefe" <raoknz(a)gmail.com> > To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> > Subject: [Pharo-users] FFI beginner question > Message-ID: > <CABcYAd+DggCnoSaQr+D54Hc5= > Mb_ToRw8+o+saeEonipyMckzw(a)mail.gmail.com> > Content-Type: text/plain; charset="UTF-8" > > I am developing a Smalltalk interface to an existing C library which I > intend to make generally available. Naturally I am doing this in my own > Smalltalk system first, where it's unsurprisingly easy for me. But when > I have it working, I'd like to make a Pharo port available. > > I have never used the Foreign Function Interface in Pharo before and > don't even know where to start. > > What should I read first? > Is there a model project to imitate? > > > > ------------------------------ > > Message: 2 > Date: Mon, 23 Sep 2019 12:40:46 +1200 > From: "Richard O'Keefe" <raoknz(a)gmail.com> > To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> > Subject: Re: [Pharo-users] Code of Conduct > Message-ID: > < > CABcYAd+Xgn45++FKgDogZgEGzQfpPCg7+ZY4t+RaPxYSYLgctQ(a)mail.gmail.com> > Content-Type: text/plain; charset="UTF-8" > > This is not a question of left vs right. It's a question of > authoritarian vs libertarian. > And this is very relevant to the community. > It's also not a question of democracy vs central authority. > It's a question of vs ???????? vs goodspeak. > And this is very relevant to the community also. > > Pharo is "owned" by the people who do the bulk of the work on it, > and they are kind enough to share it with us. That there is such a > thing as a *Pharo* community is the result of their work. > > That there is such a thing as a Pharo *community* depends on the ability of > that community to communicate freely. This cuts BOTH ways. If people are > scared off by incivility, that's bad. If people are driven away by > incivility, > that's bad. But when you stop seeing rudeness as rudeness, which may be > amended, and start seeing it as crimethink, you drive people away, and that > is bad too. > > Let's consider a recent thread. I took the position that << and putOn: > were > confusing, unreliable, and unnecessary. The unreliability issue has been > addressed in Pharo 8; had I not been able to speak I would never have > learned > that. Some people apparently think that it improves readability, where I > find > that << impairs my ability to understand. The fact that BOTH sides were > able > to speak freely means that we now know (a) that there is no consensus for > removing them from the system and (b) if you want other people to read your > code you might want to think twice before using them, and we are all > better off. > But if criticising someone's opinion were construed as harassment, the > thread > would have been shut down before I displayed my code with a generalisation > that is worth having if << is worth having at all. > > I probably should have mentioned the Erlang code of conduct > http://erlang.org/download/erlang_org_code_of_conduct.txt > It is pretty a-political, has graduated response, and potential for > forgiveness. > > A code of conduct for *events* is another matter, which is why I bring > Erlang up. > http://erlang.org/pipermail/erlang-questions/2015-March/083849.html > is eye-opening. (It's mainly about Ruby community issues.) > > On Mon, 23 Sep 2019 at 11:51, Offray Vladimir Luna C?rdenas > <offray.luna(a)mutabit.com> wrote: > > > > My point was that this community, as a the big majority of FLOSS ones, > is not a democracy and *not* having a democracy has shown its benefits in > human endeavors like science, technology, hackerspaces and so on. > > > > I'll keep the rest of the conversation with you on the source code > repository and the PR. See you there. > > > > > > On 22/09/19 6:40 p. m., Steve Quezadas wrote: > > > > This isn't science, this is a community. We don't need a CoC, there > haven't been any problems on this list regarding nazis or whatever. This is > just a group of people trying to enforce their political ideologies on > everyone else. Let's just remove the CoC altogether and just replace it > with one line: "this maillist is about Pharo, anything else is offtopic". > > > > If you want to debate on the merits of Islam vs Christianity/ right vs > left / thugs vs racists , you are free to hold your opinion on some other > sub, but it's offtopic here. > > > > On Sun, Sep 22, 2019 at 4:23 PM Offray Vladimir Luna C?rdenas < > offray.luna(a)mutabit.com> wrote: > >> > >> There is no data to support such supposed majority. But even so, free, > libre, open source communities are not democracies. Imagine the quality of > code or argumentation based on perceived majorities? If science would be a > democracy, the earth would be "still" flat. > >> > >> On 22/09/19 6:04 p. m., Steve Quezadas wrote: > >> > >> I would say that the majority don't seem to be in favor of it. This > should be a democracy. > >> > >> On Sun, Sep 22, 2019 at 1:53 PM Offray Vladimir Luna C?rdenas < > offray.luna(a)mutabit.com> wrote: > >>> > >>> > >>> On 22/09/19 3:38 p. m., Steve Quezadas wrote: > >>> > > The discussion so far shows that CoC is not a distraction to many > >>> > > >>> > Actually, the discussion shows that the CoC is "a distraction to > many". > >>> > >>> Actually it shows that some people consider it a distraction, others > >>> don't. I think that every body here is able to form its own opinion on > >>> that and invest time and effort accordingly. > >>> > >>> Cheers, > >>> > >>> Offray > >>> > >>> > >>> > > > > ------------------------------ > > Message: 3 > Date: Sun, 22 Sep 2019 19:41:09 -0500 (CDT) > From: Brainstorms <wild.ideas(a)gmail.com> > To: pharo-users(a)lists.pharo.org > Subject: Re: [Pharo-users] FFI beginner question > Message-ID: <1569199269309-0.post(a)n4.nabble.com> > Content-Type: text/plain; charset=us-ascii > > I'm also interested in this... > > Any plans to draft a Pharo booklet on this subject? > > -Ted > > > > > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html > > > > ------------------------------ > > Message: 4 > Date: Sun, 22 Sep 2019 21:37:19 -0500 > From: Offray Vladimir Luna C?rdenas <offray.luna(a)mutabit.com> > To: pharo-users(a)lists.pharo.org > Subject: Re: [Pharo-users] Code of Conduct > Message-ID: <ded9afcf-a557-9a0f-1aac-e45262b36d1a(a)mutabit.com> > Content-Type: text/plain; charset=utf-8 > > I agreed that the last decision should be on the ones who made the bulk > of the work. But I don't see relationship between a code of conduct and > not being able to talk about code or contributions quality. Just looking > at the FAQ of the original CoC that originated the whole think, I see a > lot of answers about the stuff being said on this thread (minorities, > left wing progressive agenda, diminish of code quality because of it, > mixing tech with non-tech stuff), so I will refer to it, because as I > said, I think that the PR should be the place for the bulk of the > discussion: > > https://www.contributor-covenant.org/faq > > The FAQ name goes pretty well, considering the amount of repeated > arguments they deal with. I think that many of the FAQ apply for other > CoCs, despite of the possible different nature of CoC for the online > community and the CoC for other face to face events. BTW, Thanks for the > links, both provide a better context for the emergence of the CoC in the > Erlang community. > > As said, I will try to see for specific contributions in the > correspondent PR in the repo, and made some if I have a one. For the > moment I'm trying to make my contributions on this thread, but is taking > a lot. > > Cheers, > > Offray > > On 22/09/19 7:40 p.?m., Richard O'Keefe wrote: > > This is not a question of left vs right. It's a question of > > authoritarian vs libertarian. > > And this is very relevant to the community. > > It's also not a question of democracy vs central authority. > > It's a question of vs ???????? vs goodspeak. > > And this is very relevant to the community also. > > > > Pharo is "owned" by the people who do the bulk of the work on it, > > and they are kind enough to share it with us. That there is such a > > thing as a *Pharo* community is the result of their work. > > > > That there is such a thing as a Pharo *community* depends on the ability > of > > that community to communicate freely. This cuts BOTH ways. If people > are > > scared off by incivility, that's bad. If people are driven away by > incivility, > > that's bad. But when you stop seeing rudeness as rudeness, which may be > > amended, and start seeing it as crimethink, you drive people away, and > that > > is bad too. > > > > Let's consider a recent thread. I took the position that << and putOn: > were > > confusing, unreliable, and unnecessary. The unreliability issue has been > > addressed in Pharo 8; had I not been able to speak I would never have > learned > > that. Some people apparently think that it improves readability, where > I find > > that << impairs my ability to understand. The fact that BOTH sides were > able > > to speak freely means that we now know (a) that there is no consensus for > > removing them from the system and (b) if you want other people to read > your > > code you might want to think twice before using them, and we are all > better off. > > But if criticising someone's opinion were construed as harassment, the > thread > > would have been shut down before I displayed my code with a > generalisation > > that is worth having if << is worth having at all. > > > > I probably should have mentioned the Erlang code of conduct > > http://erlang.org/download/erlang_org_code_of_conduct.txt > > It is pretty a-political, has graduated response, and potential for > forgiveness. > > > > A code of conduct for *events* is another matter, which is why I bring > > Erlang up. > > http://erlang.org/pipermail/erlang-questions/2015-March/083849.html > > is eye-opening. (It's mainly about Ruby community issues.) > > > > On Mon, 23 Sep 2019 at 11:51, Offray Vladimir Luna C?rdenas > > <offray.luna(a)mutabit.com> wrote: > >> My point was that this community, as a the big majority of FLOSS ones, > is not a democracy and *not* having a democracy has shown its benefits in > human endeavors like science, technology, hackerspaces and so on. > >> > >> I'll keep the rest of the conversation with you on the source code > repository and the PR. See you there. > >> > >> > >> On 22/09/19 6:40 p. m., Steve Quezadas wrote: > >> > >> This isn't science, this is a community. We don't need a CoC, there > haven't been any problems on this list regarding nazis or whatever. This is > just a group of people trying to enforce their political ideologies on > everyone else. Let's just remove the CoC altogether and just replace it > with one line: "this maillist is about Pharo, anything else is offtopic". > >> > >> If you want to debate on the merits of Islam vs Christianity/ right vs > left / thugs vs racists , you are free to hold your opinion on some other > sub, but it's offtopic here. > >> > >> On Sun, Sep 22, 2019 at 4:23 PM Offray Vladimir Luna C?rdenas < > offray.luna(a)mutabit.com> wrote: > >>> There is no data to support such supposed majority. But even so, free, > libre, open source communities are not democracies. Imagine the quality of > code or argumentation based on perceived majorities? If science would be a > democracy, the earth would be "still" flat. > >>> > >>> On 22/09/19 6:04 p. m., Steve Quezadas wrote: > >>> > >>> I would say that the majority don't seem to be in favor of it. This > should be a democracy. > >>> > >>> On Sun, Sep 22, 2019 at 1:53 PM Offray Vladimir Luna C?rdenas < > offray.luna(a)mutabit.com> wrote: > >>>> > >>>> On 22/09/19 3:38 p. m., Steve Quezadas wrote: > >>>>>> The discussion so far shows that CoC is not a distraction to many > >>>>> Actually, the discussion shows that the CoC is "a distraction to > many". > >>>> Actually it shows that some people consider it a distraction, others > >>>> don't. I think that every body here is able to form its own opinion on > >>>> that and invest time and effort accordingly. > >>>> > >>>> Cheers, > >>>> > >>>> Offray > >>>> > >>>> > >>>> > > > > > > > ------------------------------ > > Message: 5 > Date: Sun, 22 Sep 2019 20:14:26 -0700 > From: Steve Quezadas <steveeq1(a)gmail.com> > To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> > Subject: Re: [Pharo-users] Code of Conduct > Message-ID: > <CAJzdPQV0YZXWePmfJ_PNbmBZeWq_Ov-St1iG9h8hd4dx=9CN= > w(a)mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > > But the low rate at which marginalized people are recruited, and > > the high rate at which they leave the industry > <https://www.kaporcenter.org/tech-leavers/>, point to a larger > > cultural and systemic problem. > > Your interpreting this information with a SJW lens. Otherwise known as > "confirmation bias". Look at the low proportion of blacks and women who > apply for CS majors in college. Are you going to say that colleges are > using discriminatory practices to keep blacks and women from taking CS > classes? Maybe the bulk of the low recruitment statistics is simply due to > non-interest within that sub-culture. > > I believe this CoC is a way to wedge left-wing politics in a non-political > maillist. I want it out. > > On Sun, Sep 22, 2019 at 7:37 PM Offray Vladimir Luna C?rdenas < > offray.luna(a)mutabit.com> wrote: > > > I agreed that the last decision should be on the ones who made the bulk > > of the work. But I don't see relationship between a code of conduct and > > not being able to talk about code or contributions quality. Just looking > > at the FAQ of the original CoC that originated the whole think, I see a > > lot of answers about the stuff being said on this thread (minorities, > > left wing progressive agenda, diminish of code quality because of it, > > mixing tech with non-tech stuff), so I will refer to it, because as I > > said, I think that the PR should be the place for the bulk of the > > discussion: > > > > https://www.contributor-covenant.org/faq > > > > The FAQ name goes pretty well, considering the amount of repeated > > arguments they deal with. I think that many of the FAQ apply for other > > CoCs, despite of the possible different nature of CoC for the online > > community and the CoC for other face to face events. BTW, Thanks for the > > links, both provide a better context for the emergence of the CoC in the > > Erlang community. > > > > As said, I will try to see for specific contributions in the > > correspondent PR in the repo, and made some if I have a one. For the > > moment I'm trying to make my contributions on this thread, but is taking > > a lot. > > > > Cheers, > > > > Offray > > > > On 22/09/19 7:40 p. m., Richard O'Keefe wrote: > > > This is not a question of left vs right. It's a question of > > > authoritarian vs libertarian. > > > And this is very relevant to the community. > > > It's also not a question of democracy vs central authority. > > > It's a question of vs ???????? vs goodspeak. > > > And this is very relevant to the community also. > > > > > > Pharo is "owned" by the people who do the bulk of the work on it, > > > and they are kind enough to share it with us. That there is such a > > > thing as a *Pharo* community is the result of their work. > > > > > > That there is such a thing as a Pharo *community* depends on the > ability > > of > > > that community to communicate freely. This cuts BOTH ways. If people > > are > > > scared off by incivility, that's bad. If people are driven away by > > incivility, > > > that's bad. But when you stop seeing rudeness as rudeness, which may > be > > > amended, and start seeing it as crimethink, you drive people away, and > > that > > > is bad too. > > > > > > Let's consider a recent thread. I took the position that << and putOn: > > were > > > confusing, unreliable, and unnecessary. The unreliability issue has > been > > > addressed in Pharo 8; had I not been able to speak I would never have > > learned > > > that. Some people apparently think that it improves readability, where > > I find > > > that << impairs my ability to understand. The fact that BOTH sides > were > > able > > > to speak freely means that we now know (a) that there is no consensus > for > > > removing them from the system and (b) if you want other people to read > > your > > > code you might want to think twice before using them, and we are all > > better off. > > > But if criticising someone's opinion were construed as harassment, the > > thread > > > would have been shut down before I displayed my code with a > > generalisation > > > that is worth having if << is worth having at all. > > > > > > I probably should have mentioned the Erlang code of conduct > > > http://erlang.org/download/erlang_org_code_of_conduct.txt > > > It is pretty a-political, has graduated response, and potential for > > forgiveness. > > > > > > A code of conduct for *events* is another matter, which is why I bring > > > Erlang up. > > > http://erlang.org/pipermail/erlang-questions/2015-March/083849.html > > > is eye-opening. (It's mainly about Ruby community issues.) > > > > > > On Mon, 23 Sep 2019 at 11:51, Offray Vladimir Luna C?rdenas > > > <offray.luna(a)mutabit.com> wrote: > > >> My point was that this community, as a the big majority of FLOSS ones, > > is not a democracy and *not* having a democracy has shown its benefits in > > human endeavors like science, technology, hackerspaces and so on. > > >> > > >> I'll keep the rest of the conversation with you on the source code > > repository and the PR. See you there. > > >> > > >> > > >> On 22/09/19 6:40 p. m., Steve Quezadas wrote: > > >> > > >> This isn't science, this is a community. We don't need a CoC, there > > haven't been any problems on this list regarding nazis or whatever. This > is > > just a group of people trying to enforce their political ideologies on > > everyone else. Let's just remove the CoC altogether and just replace it > > with one line: "this maillist is about Pharo, anything else is offtopic". > > >> > > >> If you want to debate on the merits of Islam vs Christianity/ right vs > > left / thugs vs racists , you are free to hold your opinion on some other > > sub, but it's offtopic here. > > >> > > >> On Sun, Sep 22, 2019 at 4:23 PM Offray Vladimir Luna C?rdenas < > > offray.luna(a)mutabit.com> wrote: > > >>> There is no data to support such supposed majority. But even so, > free, > > libre, open source communities are not democracies. Imagine the quality > of > > code or argumentation based on perceived majorities? If science would be > a > > democracy, the earth would be "still" flat. > > >>> > > >>> On 22/09/19 6:04 p. m., Steve Quezadas wrote: > > >>> > > >>> I would say that the majority don't seem to be in favor of it. This > > should be a democracy. > > >>> > > >>> On Sun, Sep 22, 2019 at 1:53 PM Offray Vladimir Luna C?rdenas < > > offray.luna(a)mutabit.com> wrote: > > >>>> > > >>>> On 22/09/19 3:38 p. m., Steve Quezadas wrote: > > >>>>>> The discussion so far shows that CoC is not a distraction to many > > >>>>> Actually, the discussion shows that the CoC is "a distraction to > > many". > > >>>> Actually it shows that some people consider it a distraction, others > > >>>> don't. I think that every body here is able to form its own opinion > on > > >>>> that and invest time and effort accordingly. > > >>>> > > >>>> Cheers, > > >>>> > > >>>> Offray > > >>>> > > >>>> > > >>>> > > > > > > > > > >

Sept. 23, 2019

Re: [Pharo-users] FFI beginner question

by TomaÅ¾ Turk

Hi, The book draft is here: https://files.pharo.org/books-pdfs/booklet-uFFI/UFFIDRAFT.pdf. Best wishes, Tomaz > >Any plans to draft a Pharo booklet on this subject? > >-Ted >

Sept. 23, 2019

Re: [Pharo-users] Code of Conduct

by Steve Quezadas

> But the low rate at which marginalized people are recruited, and > the high rate at which they leave the industry <https://www.kaporcenter.org/tech-leavers/>, point to a larger > cultural and systemic problem. Your interpreting this information with a SJW lens. Otherwise known as "confirmation bias". Look at the low proportion of blacks and women who apply for CS majors in college. Are you going to say that colleges are using discriminatory practices to keep blacks and women from taking CS classes? Maybe the bulk of the low recruitment statistics is simply due to non-interest within that sub-culture. I believe this CoC is a way to wedge left-wing politics in a non-political maillist. I want it out. On Sun, Sep 22, 2019 at 7:37 PM Offray Vladimir Luna CÃ¡rdenas < offray.luna(a)mutabit.com> wrote: > I agreed that the last decision should be on the ones who made the bulk > of the work. But I don't see relationship between a code of conduct and > not being able to talk about code or contributions quality. Just looking > at the FAQ of the original CoC that originated the whole think, I see a > lot of answers about the stuff being said on this thread (minorities, > left wing progressive agenda, diminish of code quality because of it, > mixing tech with non-tech stuff), so I will refer to it, because as I > said, I think that the PR should be the place for the bulk of the > discussion: > > https://www.contributor-covenant.org/faq > > The FAQ name goes pretty well, considering the amount of repeated > arguments they deal with. I think that many of the FAQ apply for other > CoCs, despite of the possible different nature of CoC for the online > community and the CoC for other face to face events. BTW, Thanks for the > links, both provide a better context for the emergence of the CoC in the > Erlang community. > > As said, I will try to see for specific contributions in the > correspondent PR in the repo, and made some if I have a one. For the > moment I'm trying to make my contributions on this thread, but is taking > a lot. > > Cheers, > > Offray > > On 22/09/19 7:40 p. m., Richard O'Keefe wrote: > > This is not a question of left vs right. It's a question of > > authoritarian vs libertarian. > > And this is very relevant to the community. > > It's also not a question of democracy vs central authority. > > It's a question of vs ÏÎ±ÏÏÎ·ÏÎ¯Î± vs goodspeak. > > And this is very relevant to the community also. > > > > Pharo is "owned" by the people who do the bulk of the work on it, > > and they are kind enough to share it with us. That there is such a > > thing as a *Pharo* community is the result of their work. > > > > That there is such a thing as a Pharo *community* depends on the ability > of > > that community to communicate freely. This cuts BOTH ways. If people > are > > scared off by incivility, that's bad. If people are driven away by > incivility, > > that's bad. But when you stop seeing rudeness as rudeness, which may be > > amended, and start seeing it as crimethink, you drive people away, and > that > > is bad too. > > > > Let's consider a recent thread. I took the position that << and putOn: > were > > confusing, unreliable, and unnecessary. The unreliability issue has been > > addressed in Pharo 8; had I not been able to speak I would never have > learned > > that. Some people apparently think that it improves readability, where > I find > > that << impairs my ability to understand. The fact that BOTH sides were > able > > to speak freely means that we now know (a) that there is no consensus for > > removing them from the system and (b) if you want other people to read > your > > code you might want to think twice before using them, and we are all > better off. > > But if criticising someone's opinion were construed as harassment, the > thread > > would have been shut down before I displayed my code with a > generalisation > > that is worth having if << is worth having at all. > > > > I probably should have mentioned the Erlang code of conduct > > http://erlang.org/download/erlang_org_code_of_conduct.txt > > It is pretty a-political, has graduated response, and potential for > forgiveness. > > > > A code of conduct for *events* is another matter, which is why I bring > > Erlang up. > > http://erlang.org/pipermail/erlang-questions/2015-March/083849.html > > is eye-opening. (It's mainly about Ruby community issues.) > > > > On Mon, 23 Sep 2019 at 11:51, Offray Vladimir Luna CÃ¡rdenas > > <offray.luna(a)mutabit.com> wrote: > >> My point was that this community, as a the big majority of FLOSS ones, > is not a democracy and *not* having a democracy has shown its benefits in > human endeavors like science, technology, hackerspaces and so on. > >> > >> I'll keep the rest of the conversation with you on the source code > repository and the PR. See you there. > >> > >> > >> On 22/09/19 6:40 p. m., Steve Quezadas wrote: > >> > >> This isn't science, this is a community. We don't need a CoC, there > haven't been any problems on this list regarding nazis or whatever. This is > just a group of people trying to enforce their political ideologies on > everyone else. Let's just remove the CoC altogether and just replace it > with one line: "this maillist is about Pharo, anything else is offtopic". > >> > >> If you want to debate on the merits of Islam vs Christianity/ right vs > left / thugs vs racists , you are free to hold your opinion on some other > sub, but it's offtopic here. > >> > >> On Sun, Sep 22, 2019 at 4:23 PM Offray Vladimir Luna CÃ¡rdenas < > offray.luna(a)mutabit.com> wrote: > >>> There is no data to support such supposed majority. But even so, free, > libre, open source communities are not democracies. Imagine the quality of > code or argumentation based on perceived majorities? If science would be a > democracy, the earth would be "still" flat. > >>> > >>> On 22/09/19 6:04 p. m., Steve Quezadas wrote: > >>> > >>> I would say that the majority don't seem to be in favor of it. This > should be a democracy. > >>> > >>> On Sun, Sep 22, 2019 at 1:53 PM Offray Vladimir Luna CÃ¡rdenas < > offray.luna(a)mutabit.com> wrote: > >>>> > >>>> On 22/09/19 3:38 p. m., Steve Quezadas wrote: > >>>>>> The discussion so far shows that CoC is not a distraction to many > >>>>> Actually, the discussion shows that the CoC is "a distraction to > many". > >>>> Actually it shows that some people consider it a distraction, others > >>>> don't. I think that every body here is able to form its own opinion on > >>>> that and invest time and effort accordingly. > >>>> > >>>> Cheers, > >>>> > >>>> Offray > >>>> > >>>> > >>>> > > > > >

Sept. 23, 2019