- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] TxText model

by Igor Stasenko

On 4 April 2016 at 15:44, Thierry Goubier <thierry.goubier(a)gmail.com> wrote: > > > Have you really looked into what is the core of the TeX algorithm? The > fact that an interactive version of it was done multiple times in history? > (Self / InterViews to cite the ones I know and have used) > > Nope. Text processing was never on radar of my interests. And fact that we can reproduce same functionality in our environment won't bring us anywhere closer to LaTex or any other word-processing beasts. As i said, being able to make rocket engine doesn't means you can travel to the Moon. Or Mars. Yes, it is minimal requirement.. but just one of many. -- Best regards, Igor Stasenko.

April 4, 2016

Re: [Pharo-dev] TxText model

by Igor Stasenko

On 4 April 2016 at 15:44, Thierry Goubier <thierry.goubier(a)gmail.com> wrote: > > > 2016-04-04 14:24 GMT+02:00 Igor Stasenko <siguctua(a)gmail.com>: > >> >> >> On 4 April 2016 at 14:28, Thierry Goubier <thierry.goubier(a)gmail.com> >> wrote: >> >>> >>> >>> 2016-04-04 13:18 GMT+02:00 Stephan Eggermont <stephan(a)stack.nl>: >>> >>>> On 04-04-16 11:58, Igor Stasenko wrote: >>>> >>>>> Apart from being 'cool to have', full-fledged word processing is not a >>>>> thing, that you dealing with on a daily basis in environment, like Pharo. >>>>> >>>> I'm sure that is the case for you. I wonder if that is the case for >>>> many Pharo users. >>>> AFAIK there are a lot of pillar users. >>>> >>> >>> If it is for Pillar, then you don't really need a full-fledged, paper >>> oriented layout engine. A web-like layout environment is probably enough, >>> and much less costly to build. >>> >>> >> Now count, how much world-wide resources are dedicated to web-based and >> browser-based technology development and compare with our resource base. I >> think it is foolish to set an unrealistic goals. >> > > I think that if we didn't have unrealistic goals, we wouldn't be in that > community :) > > >> >>> For me the problem with the TxText model is that it blocks the >>>> possibility of doing >>>> that later, if and when there is enough development capacity to invest >>>> in this. >>> >>> >>> There is enough technology in the Pharo universe to do it (or at least >>> something approaching). Sometimes, what you need is the ideas / the >>> rationale from a project to do it. And I do believe TxText has some of it, >>> even if you consider that TxText can't be extended to do it (and I'll >>> consider that you are right on this). >>> >>> Now, it's on nobody's roadmap, so it may take a while to emerge (if it >>> does at all). >>> >>> Last time, i installed LaTex package on my mac, it took maybe hour or >> so.. About 1Gb of files, tools, compilers, GUI, text editors.. >> Now think, how much years it would take to get remotely close to such >> level of development? And where are those people or money that would allow >> us to think this is viable path and we should throw everything into it to >> get there? >> > > Have you really looked into what is the core of the TeX algorithm? The > fact that an interactive version of it was done multiple times in history? > (Self / InterViews to cite the ones I know and have used) > > >> It is nice to dream time to time, but let us be realistic. >> > > I'd say that we have an environment where we can dream; where reinventions > can be done at low cost (or at lower cost than others), that we can even > afford to be wrong and throw a project away. Otherwise we wouldn't have > Athens / TxText / GT ... > > Mind you, I'm not asking you to do it :) I simply know that some of TxText > stuff is usable in that context. And that given the goal, shortcuts are > possible. > > But do you realising that we talking here about different scales of things? Let me drive an analogy: - you found an engineer that created a rocket engine in his garage. Engine is perfect, stable, works well etc etc etc.. And you asking him: - can we fly to the Moon tomorrow? Huh? Okay, if you find specialist, who can plan the mission, find the producer of solar panels, find specialist of long range communications, find specialists of long-range observations to determine the landing site, find god know how many other specialist and experts in various areas, only then you could possibly find and answer to your question. But asking such question to just a rocket engine specialist.. is just foolish. > Regards, > > Thierry > -- Best regards, Igor Stasenko.

April 4, 2016

Re: [Pharo-dev] TxText model

by Thierry Goubier

2016-04-04 14:49 GMT+02:00 Igor Stasenko <siguctua(a)gmail.com>: > Oh, and aside all of that.. Making a full-fledged word processor is not > just a regular engineering task. You need an expert of publishing, expert > in fonts and typography. That's right from the beginning. > That's why I said that I believe the shortcut is to not focus on the paper-like; for pillar, just use a pdf converter (latex output or whatever). > And i am not that expert in this domain(s). > Neither am I. > So, next time, when we start talking about things like page layouts, > columns, margins, tabs, references and other stuff, first find an expert > who will be able to transform these terms into technical requirements. > If you think it so simple, it is not: because all those terms came from > paper-publishing domain, that existed even before first computer came to > existence. > The capabilities of TxText and whether it will be capable to handle well > complexities of full-fledged word processing features, at this point, > without having an expert is nothing but just a speculation. > And a lot of time... agreed. Thierry

April 4, 2016

Re: [Pharo-dev] TxText model

by Igor Stasenko

Oh, and aside all of that.. Making a full-fledged word processor is not just a regular engineering task. You need an expert of publishing, expert in fonts and typography. That's right from the beginning. And i am not that expert in this domain(s). So, next time, when we start talking about things like page layouts, columns, margins, tabs, references and other stuff, first find an expert who will be able to transform these terms into technical requirements. If you think it so simple, it is not: because all those terms came from paper-publishing domain, that existed even before first computer came to existence. The capabilities of TxText and whether it will be capable to handle well complexities of full-fledged word processing features, at this point, without having an expert is nothing but just a speculation. On 4 April 2016 at 15:24, Igor Stasenko <siguctua(a)gmail.com> wrote: > > > On 4 April 2016 at 14:28, Thierry Goubier <thierry.goubier(a)gmail.com> > wrote: > >> >> >> 2016-04-04 13:18 GMT+02:00 Stephan Eggermont <stephan(a)stack.nl>: >> >>> On 04-04-16 11:58, Igor Stasenko wrote: >>> >>>> Apart from being 'cool to have', full-fledged word processing is not a >>>> thing, that you dealing with on a daily basis in environment, like Pharo. >>>> >>> I'm sure that is the case for you. I wonder if that is the case for many >>> Pharo users. >>> AFAIK there are a lot of pillar users. >>> >> >> If it is for Pillar, then you don't really need a full-fledged, paper >> oriented layout engine. A web-like layout environment is probably enough, >> and much less costly to build. >> >> > Now count, how much world-wide resources are dedicated to web-based and > browser-based technology development and compare with our resource base. I > think it is foolish to set an unrealistic goals. > > > >> For me the problem with the TxText model is that it blocks the >>> possibility of doing >>> that later, if and when there is enough development capacity to invest >>> in this. >> >> >> There is enough technology in the Pharo universe to do it (or at least >> something approaching). Sometimes, what you need is the ideas / the >> rationale from a project to do it. And I do believe TxText has some of it, >> even if you consider that TxText can't be extended to do it (and I'll >> consider that you are right on this). >> >> Now, it's on nobody's roadmap, so it may take a while to emerge (if it >> does at all). >> >> Last time, i installed LaTex package on my mac, it took maybe hour or > so.. About 1Gb of files, tools, compilers, GUI, text editors.. > Now think, how much years it would take to get remotely close to such > level of development? And where are those people or money that would allow > us to think this is viable path and we should throw everything into it to > get there? > It is nice to dream time to time, but let us be realistic. > > >> Thierry >> > > > > -- > Best regards, > Igor Stasenko. > -- Best regards, Igor Stasenko.

April 4, 2016

Re: [Pharo-dev] TxText model

by Thierry Goubier

2016-04-04 14:14 GMT+02:00 Igor Stasenko <siguctua(a)gmail.com>: > > > On 4 April 2016 at 14:18, Stephan Eggermont <stephan(a)stack.nl> wrote: > >> On 04-04-16 11:58, Igor Stasenko wrote: >> >>> Apart from being 'cool to have', full-fledged word processing is not a >>> thing, that you dealing with on a daily basis in environment, like Pharo. >>> >> I'm sure that is the case for you. I wonder if that is the case for many >> Pharo users. >> AFAIK there are a lot of pillar users. >> >> AFAIK there are a lot of car drivers.. does it means Pharo should focus > development towards needs of car drivers? :) > > >> For me the problem with the TxText model is that it blocks the >> possibility of doing >> that later, if and when there is enough development capacity to invest in >> this. > > > Why you have such impression? I just said that primary focus was to > deliver model that works fine for our current needs. For something more > sophisticated simply we had no resources. Take that into account. > > I had no focus or plans for turning model into full-fledged word > processor in future. Right. But i don't see how the model could block such > development, if such need would arise at some point. > > If you have doubts about model, feel free to point it out and we can > discuss details. > I was hoping for that sentence ! Regards, Thierry > >> >> Stephan >> >> > > -- > Best regards, > Igor Stasenko. >

April 4, 2016

Re: [Pharo-dev] TxText model

by Thierry Goubier

2016-04-04 14:24 GMT+02:00 Igor Stasenko <siguctua(a)gmail.com>: > > > On 4 April 2016 at 14:28, Thierry Goubier <thierry.goubier(a)gmail.com> > wrote: > >> >> >> 2016-04-04 13:18 GMT+02:00 Stephan Eggermont <stephan(a)stack.nl>: >> >>> On 04-04-16 11:58, Igor Stasenko wrote: >>> >>>> Apart from being 'cool to have', full-fledged word processing is not a >>>> thing, that you dealing with on a daily basis in environment, like Pharo. >>>> >>> I'm sure that is the case for you. I wonder if that is the case for many >>> Pharo users. >>> AFAIK there are a lot of pillar users. >>> >> >> If it is for Pillar, then you don't really need a full-fledged, paper >> oriented layout engine. A web-like layout environment is probably enough, >> and much less costly to build. >> >> > Now count, how much world-wide resources are dedicated to web-based and > browser-based technology development and compare with our resource base. I > think it is foolish to set an unrealistic goals. > I think that if we didn't have unrealistic goals, we wouldn't be in that community :) > >> For me the problem with the TxText model is that it blocks the >>> possibility of doing >>> that later, if and when there is enough development capacity to invest >>> in this. >> >> >> There is enough technology in the Pharo universe to do it (or at least >> something approaching). Sometimes, what you need is the ideas / the >> rationale from a project to do it. And I do believe TxText has some of it, >> even if you consider that TxText can't be extended to do it (and I'll >> consider that you are right on this). >> >> Now, it's on nobody's roadmap, so it may take a while to emerge (if it >> does at all). >> >> Last time, i installed LaTex package on my mac, it took maybe hour or > so.. About 1Gb of files, tools, compilers, GUI, text editors.. > Now think, how much years it would take to get remotely close to such > level of development? And where are those people or money that would allow > us to think this is viable path and we should throw everything into it to > get there? > Have you really looked into what is the core of the TeX algorithm? The fact that an interactive version of it was done multiple times in history? (Self / InterViews to cite the ones I know and have used) > It is nice to dream time to time, but let us be realistic. > I'd say that we have an environment where we can dream; where reinventions can be done at low cost (or at lower cost than others), that we can even afford to be wrong and throw a project away. Otherwise we wouldn't have Athens / TxText / GT ... Mind you, I'm not asking you to do it :) I simply know that some of TxText stuff is usable in that context. And that given the goal, shortcuts are possible. Regards, Thierry

April 4, 2016

Re: [Pharo-dev] TxText model

by Igor Stasenko

On 4 April 2016 at 14:28, Thierry Goubier <thierry.goubier(a)gmail.com> wrote: > > > 2016-04-04 13:18 GMT+02:00 Stephan Eggermont <stephan(a)stack.nl>: > >> On 04-04-16 11:58, Igor Stasenko wrote: >> >>> Apart from being 'cool to have', full-fledged word processing is not a >>> thing, that you dealing with on a daily basis in environment, like Pharo. >>> >> I'm sure that is the case for you. I wonder if that is the case for many >> Pharo users. >> AFAIK there are a lot of pillar users. >> > > If it is for Pillar, then you don't really need a full-fledged, paper > oriented layout engine. A web-like layout environment is probably enough, > and much less costly to build. > > Now count, how much world-wide resources are dedicated to web-based and browser-based technology development and compare with our resource base. I think it is foolish to set an unrealistic goals. > For me the problem with the TxText model is that it blocks the possibility >> of doing >> that later, if and when there is enough development capacity to invest in >> this. > > > There is enough technology in the Pharo universe to do it (or at least > something approaching). Sometimes, what you need is the ideas / the > rationale from a project to do it. And I do believe TxText has some of it, > even if you consider that TxText can't be extended to do it (and I'll > consider that you are right on this). > > Now, it's on nobody's roadmap, so it may take a while to emerge (if it > does at all). > > Last time, i installed LaTex package on my mac, it took maybe hour or so.. About 1Gb of files, tools, compilers, GUI, text editors.. Now think, how much years it would take to get remotely close to such level of development? And where are those people or money that would allow us to think this is viable path and we should throw everything into it to get there? It is nice to dream time to time, but let us be realistic. > Thierry > -- Best regards, Igor Stasenko.

April 4, 2016

Re: [Pharo-dev] TxText model

by Igor Stasenko

On 4 April 2016 at 14:18, Stephan Eggermont <stephan(a)stack.nl> wrote: > On 04-04-16 11:58, Igor Stasenko wrote: > >> Apart from being 'cool to have', full-fledged word processing is not a >> thing, that you dealing with on a daily basis in environment, like Pharo. >> > I'm sure that is the case for you. I wonder if that is the case for many > Pharo users. > AFAIK there are a lot of pillar users. > > AFAIK there are a lot of car drivers.. does it means Pharo should focus development towards needs of car drivers? :) > For me the problem with the TxText model is that it blocks the possibility > of doing > that later, if and when there is enough development capacity to invest in > this. Why you have such impression? I just said that primary focus was to deliver model that works fine for our current needs. For something more sophisticated simply we had no resources. Take that into account. I had no focus or plans for turning model into full-fledged word processor in future. Right. But i don't see how the model could block such development, if such need would arise at some point. If you have doubts about model, feel free to point it out and we can discuss details. > > Stephan > > -- Best regards, Igor Stasenko.

April 4, 2016

Re: [Pharo-dev] TxText model

by Thierry Goubier

2016-04-04 13:18 GMT+02:00 Stephan Eggermont <stephan(a)stack.nl>: > On 04-04-16 11:58, Igor Stasenko wrote: > >> Apart from being 'cool to have', full-fledged word processing is not a >> thing, that you dealing with on a daily basis in environment, like Pharo. >> > I'm sure that is the case for you. I wonder if that is the case for many > Pharo users. > AFAIK there are a lot of pillar users. > If it is for Pillar, then you don't really need a full-fledged, paper oriented layout engine. A web-like layout environment is probably enough, and much less costly to build. > For me the problem with the TxText model is that it blocks the possibility > of doing > that later, if and when there is enough development capacity to invest in > this. There is enough technology in the Pharo universe to do it (or at least something approaching). Sometimes, what you need is the ideas / the rationale from a project to do it. And I do believe TxText has some of it, even if you consider that TxText can't be extended to do it (and I'll consider that you are right on this). Now, it's on nobody's roadmap, so it may take a while to emerge (if it does at all). Thierry

April 4, 2016

TxText model

by Stephan Eggermont

On 04-04-16 11:58, Igor Stasenko wrote: > Apart from being 'cool to have', full-fledged word processing is not a > thing, that you dealing with on a daily basis in environment, like Pharo. I'm sure that is the case for you. I wonder if that is the case for many Pharo users. AFAIK there are a lot of pillar users. For me the problem with the TxText model is that it blocks the possibility of doing that later, if and when there is enough development capacity to invest in this. Stephan

April 4, 2016