[07:29:13] Thank you [09:23:40] Thank you [17:06:53] Aside of the fact that we use it already for a lot of things – what makes Mediawiki a particularly good platform for Wikilambada? (It would be that, right?) Wikilambada seems to be code heavvy, and I wonder if Mediwiki manages this well. [17:07:06] *Mediawiki [17:08:21] The funny thing is - it is "wiki" going back to its roots. Ward's original wiki concept was for programmers to share patterns [17:08:54] @vrandecic I wonder, do you know if Ward knows about Abstract Wikipedia plans and has he reacted to it? [17:10:10] Cunningham? He is busy with federated wiki these days no? (re @fuzheado: @vrandecic I wonder, do you know if Ward knows about Abstract Wikipedia plans and has he reacted to it?) [17:10:46] So while i don't know if I would say that Mediawiki, per se, is a particularly good platform for Wikilambda... WikiWikiWeb and the Portland Pattern Repository that inspired Wikipedia is certainly in line with what Abstract Wikipedia strives to be! (re @Jandit: Aside of the fact that we use it already for a lot of things – what makes Mediawiki a particularly good platform for Wikilambada? (It would be that, right?) Wikilamb [17:11:16] Yes, and I was actually on a call with him a few weeks ago and didn't think to ask him about Abstract Wikipedia! (re @zBlace: Cunningham? He is busy with federated wiki these days no?) [17:11:51] https://en.m.wikipedia.org/wiki/WikiWikiWeb [17:24:45] Ward have the world wonderful things, but his two best-known inventions have a similar problem. [17:24:46] His first famous invention is the wiki idea, of course. His other famous invention is much less famous than the wiki, but somewhat well-known among software developers and quality assurance engineers "Fit", or https://en.wikipedia.org/wiki/Framework_for_integrated_test. [17:24:48] The common problem to both ideas is that both try to have code without code. The wiki has markup. Ward's intention, and early Wikipedians' intention was that the markup would be super-simple, much simpler than HTML. As experience showed, wikitext became much more complex. [17:24:49] "Fit" has a similar problem. It is a way to prepare software tests by "simply" writing an HTML table that describes them in a human-readable language that managers and engineers can easily read and write, but it's also machine-readable, so it can run automated tests. I worked in a company that used it in 2007–2008. Of course, to actually make it machine-readable, the HTML table code has to be written in a certain way, so it' [17:24:50] My takeaway from this is that "code without code" is not actually possible. Whenever you find yourself thinking "it looks just like human language, but actually it also runs programs", it's not human language. It doesn't mean that this thing is bad or not useful; it may be very good. It just shouldn't be presented as "easy", "human", or "natural". Code is code. [17:24:52] Ward gave the world wonderful things, but his two best-known inventions have a similar problem. [17:24:53] His first famous invention is the wiki idea, of course. His other famous invention is much less famous than the wiki, but somewhat well-known among software developers and quality assurance engineers "Fit", or https://en.wikipedia.org/wiki/Framework_for_integrated_test. [17:24:54] The common problem to both ideas is that both try to have code without code. The wiki has markup. Ward's intention, and early Wikipedians' intention was that the markup would be super-simple, much simpler than HTML. As experience showed, wikitext became much more complex. [17:24:55] "Fit" has a similar problem. It is a way to prepare software tests by "simply" writing an HTML table that describes them in a human-readable language that managers and engineers can easily read and write, but it's also machine-readable, so it can run automated tests. I worked in a company that used it in 2007–2008. Of course, to actually make it machine-readable, the HTML table code has to be written in a certain way, so it' [17:24:57] My takeaway from this is that "code without code" is not actually possible. Whenever you find yourself thinking "it looks just like human language, but actually it also runs programs", it's not human language. It doesn't mean that this thing is bad or not useful; it may be very good. It just shouldn't be presented as "easy", "human", or "natural". Code is code. [17:25:15] Ward gave the world wonderful things, but his two best-known inventions have a similar problem. [17:25:16] His first famous invention is the wiki idea, of course. His other famous invention is much less famous than the wiki, but somewhat well-known among software developers and quality assurance engineers: "Fit", or https://en.wikipedia.org/wiki/Framework_for_integrated_test. [17:25:17] The common problem to both ideas is that both try to have code without code. The wiki has markup. Ward's intention, and early Wikipedians' intention was that the markup would be super-simple, much simpler than HTML. As experience showed, wikitext became much more complex. [17:25:19] "Fit" has a similar problem. It is a way to prepare software tests by "simply" writing an HTML table that describes them in a human-readable language that managers and engineers can easily read and write, but it's also machine-readable, so it can run automated tests. I worked in a company that used it in 2007–2008. Of course, to actually make it machine-readable, the HTML table code has to be written in a certain way, so it' [17:25:20] My takeaway from this is that "code without code" is not actually possible. Whenever you find yourself thinking "it looks just like human language, but actually it also runs programs", it's not human language. It doesn't mean that this thing is bad or not useful; it may be very good. It just shouldn't be presented as "easy", "human", or "natural". Code is code. [17:30:40] Interesting! I hadn't known much about "Fit" until you described it. [17:30:41] Though, I think we cannot lay the problem of wiki markup evolving into a large beast at the feet of Ward. After all, he was the one who told Jimmy that yes you can use a wiki to create an online encyclopedia, but that it would quickly not be a "wiki" anymore, in the purest sense (re @amire80: Ward gave the world wonderful things, but his two best-known inventions have a similar problem. [17:30:42] His first famous invention is the wiki idea, of course. His other famous invention is much less famous than the wiki, but somewhat well-known among software developers and quality assurance engineers: "Fit", or Framework for integrated test. [17:30:44] The common problem to both ideas is that both try to have code without code. The wiki has markup. Ward's intention, and early Wikipedians' intention was that the markup would be super-simple, much simpler than HTML. As experience showed, wikitext became much more complex. [17:30:45] "Fit" has a similar problem. It is a way to prepare software tests by "simply" writing an HTML table that describes them in a human-readable language that managers and engineers can easily read and write, but it's also machine-readable, so it can run automated tests. I worked in a company that used it in 2007–2008. Of course, to actually make it machine-readable, the HTML table code has to be written in a certain way, so it [17:30:46] My takeaway from this is that "code without code" is not actually possible. Whenever you find yourself thinking "it looks just like human language, but actually it also runs programs", it's not human language. It doesn't mean that this thing is bad or not useful; it may be very good. It just shouldn't be presented as "easy", "human", or "natural". Code is code.) [17:32:00] The needs of an multilingual, encyclopedia-writing community facing an onslaught of external trolls and miscreants tilted Mediawiki heavily towards certain features that probably weren't a priority in his original conception of wiki :) [17:57:46] True. Much of the huge aspirations and their repercuations were dealth in technocratic way via software, with only so little in social/cultural methods. (re @fuzheado: The needs of an multilingual, encyclopedia-writing community facing an onslaught of external trolls and miscreants tilted Mediawiki heavily towards certain features that probably weren't a priority in his original conception of wiki :)) [18:01:08] But then also...non text based options are super under-used https://youtu.be/QQhVQ1UG6aM (re @amire80: Ward gave the world wonderful things, but his two best-known inventions have a similar problem. [18:01:09] His first famous invention is the wiki idea, of course. His other famous invention is much less famous than the wiki, but somewhat well-known among software developers and quality assurance engineers: "Fit", or Framework for integrated test. [18:01:10] The common problem to both ideas is that both try to have code without code. The wiki has markup. Ward's intention, and early Wikipedians' intention was that the markup would be super-simple, much simpler than HTML. As experience showed, wikitext became much more complex. [18:01:11] "Fit" has a similar problem. It is a way to prepare software tests by "simply" writing an HTML table that describes them in a human-readable language that managers and engineers can easily read and write, but it's also machine-readable, so it can run automated tests. I worked in a company that used it in 2007–2008. Of course, to actually make it machine-readable, the HTML table code has to be written in a certain way, so it's [18:01:13] My takeaway from this is that "code without code" is not actually possible. Whenever you find yourself thinking "it looks just like human language, but actually it also runs programs", it's not human language. It doesn't mean that this thing is bad or not useful; it may be very good. It just shouldn't be presented as "easy", "human", or "natural". Code is code.) [21:33:31] So I know that Alan Kay is aware about Wikilambda, I'm not aware that Ward Cunningham is. [21:39:47] I'd be curious what Ward thinks. I'm sure he would have valuable thoughts. [22:03:01] @vrandecic btw, what is the plan for https://github.com/google/abstracttext/? I saw that development for a WikiLambda extension has started on Gerrit/Phabricator, will that replace AbstractText? [22:04:50] Still open [22:04:57] ok [22:05:55] In Gerrit so far all code is in PHP. AbstractText has most of its code in JS. We will need to figure it out when we get there. [22:06:56] for now I’ll keep hacking on GraalEneyj and hope I can repurpose most of the code for upstream changes :) [22:08:10] Yes, personally I think that I would like to rewrite AbstractText to provide a fun CLI experience