August 22, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Dave Jones | On Monday, 20 August 2018 at 08:31:15 UTC, Dave Jones wrote:
> That's what Im trying to say. Im sure posts like that are popular within the D community but they are not going to make much headway bringing new users in.
We had "D parser smokes the competition" posts.
| |||
Permalink
ReplyAugust 22, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Kagamin | On Wednesday, 22 August 2018 at 13:28:37 UTC, Kagamin wrote:
> On Monday, 20 August 2018 at 08:31:15 UTC, Dave Jones wrote:
>> That's what Im trying to say. Im sure posts like that are popular within the D community but they are not going to make much headway bringing new users in.
>
> We had "D parser smokes the competition" posts.
Unfortunately, with all the D parsers that smoked the competition, we are mostly stuck with std.xml (dxml might changed this) and std.json, because those other projects never made it into the stdlib for one reason for another (not being 100% range based, not supporting XYZ memory allocator).
| |||
August 22, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Kagamin | On Wednesday, 22 August 2018 at 13:17:00 UTC, Kagamin wrote: > On Monday, 20 August 2018 at 03:57:10 UTC, John Carter wrote: >> * Choice. ie. Programmers _want_ to use it, not are constrained to use it. >> * For programming activity, not new projects. ie. The era of vast tracts of green field programming is long gone. We're mostly in the era tinker toys and tidying. > > That's a matter of choice, some are tidying, but there's a lot of green field programming even in C, and new languages are all green fields. I suspect if you actually lean of the shoulder of the vast majority programmers earning their daily bread, they aren't writing a brand new program... they enhancing, and fixing an existing one. >> There is a big difference between "Doing a lot of" and "Being Good at". > > That's why you can't be tidying all the time, you can improve, but can't become good this way. Oh, I would argue it's the best way. Or this wouldn't be funny.... http://bonkersworld.net/building-software > >> By tidying I mean refactoring legacy code that is way too large and complex to rewrite all at once. > > Nobody is going to deep refactoring; example: C/C++ (well, you mention them too) and pretty much everything. And it's that large because it accumulated garbage and rewrite will cut it to a manageable size; example: s2n (fun fact: it's written in C, but uses slices for safety just like D). Whenever I see a rewrite which claims it has made things so wondrously simpler / better, closer inspection reveals it does wondrously less, and supports wondrously less legacy cruft. Thus I do not believe these "experiments" have isolated the effect deleting unneeded or little used features and support for legacy platforms, vs the effect of rewriting vs refactoring. > Nobody is going to deep refactoring That I believe could be the paradigm shifting advantage of D. Every time I have written a refactoring or code analysis tool for C or C++, the preprocessor has amplified the complexity of my task by orders of magnitude. And every transformation I might propose.... it is incredibly hard to guarantee that it is safe and behaviour preserving, a sentiment echo'd by every optimization pass writer for C/C++. | |||
August 23, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Andrei Alexandrescu | On Saturday, 18 August 2018 at 13:33:43 UTC, Andrei Alexandrescu wrote: > where are the best leverage points in making the D language more successful. I'm still internalizing the article and thinking about how it applies to the "D system", but I've always thought facilitating the incorporation of GDC into GCC to be the single most accelerating thing we could do to gain more adoption. It somewhat fits into *7. The gain around driving positive feedback loops*. But there's risk associated with that. Walter has often said that "build it and they will come" is a Hollywood myth, but I disagree. Part of the reason why D hasn't achieved mass adoption, isn't because it's not marketed well, but because it has a number of flaws. Most of us see the *potential* of D, and are able to look past the flaws, with the faith (hopefully not misplaced) that they will one day be addressed. Others only see the flaws and the appeal of other programming languages with more resources, better management, more talent, and especially more velocity toward their goals. I often worry that if we encourage adoption, before we have something worthy of adoption, we'll only leave users with a bad taste in their mouth [0]. I've already seen a number of people, some major contributors, leave D for greener pastures. Most of the contributors that built the D runtime and did the majority of bug fixing in the compiler are gone now. At this point in time, I can only recommend D professionally to teams that are risk takers, have the aptitude to solve their own problems, and have the resources and willingness to be D contributors. We should probably be looking more for leverage points to help us better capitalize on the resources and talent we have and bring in more. Unfortunately I'm seeing an over-correction in *8. The strength of negative feedback loops, relative to the impacts they are trying to correct against*. As we try to get contributors to focus on the things that matter (at least to the powers that be), we frustrate them until they close their pull requests or just give up [1] [2]. It took me a few years to find my "in", and I'm still not really "in", but I learned that the *little things* that some consider a distraction are how people get started contributing to D. I've often said that we actually don't need more contributors; but more reviewers. There's a catch to that, though; they're not going to become reviewers if they can't first become contributors. So perhaps, I need to correct my perspective. So, I'll close with this: We should probably be more welcoming to those willing to contribute, let them work on the little stuff that's important to them, throw them a bone or two, review their pull requests in a timely manner, etc... I think those contributors will eventually become our reviewers, and then they will eventually lessen the burden so veterans can focus on the things that they think are higher priorities. This is a positive feedback loop. Help people become positive contributors, and those contributors will eventually help the next generation. I think there are a few little things the leadership, especially, can do to prime that pump, starting with being more active, helpful, and gracious with things that are currently sitting in the PR queue. Though it's a two-way street, and some contributors could also be more cooperative also. Walter and a few others have been quite gracious to me [3] [4]. I've tried to pay that forward and help other contributors find their "in", but I'm still not able to review and make decisions about many things, so I'm only of limited help. I don't think others have been treated as well. Mike [0] - https://issues.dlang.org/show_bug.cgi?id=14100 - Link in that issue no longer exists, but let's just say the user wasn't happy with D [1] - https://github.com/dlang/dmd/pulls?q=is%3Apr+author%3Amarler8997+is%3Aclosed [2] - https://github.com/dlang/dmd/pull/8378 [3] - https://github.com/dlang/dmd/pull/7395#issuecomment-349200847 [4] - https://github.com/dlang/dmd/pull/7055#issuecomment-320006283 | |||
August 24, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Andrei Alexandrescu | On Saturday, 18 August 2018 at 13:33:43 UTC, Andrei Alexandrescu wrote: > A friend recommended this article: > > http://donellameadows.org/archives/leverage-points-places-to-intervene-in-a-system/ > > I found it awesome and would recommend to anyone in this community. Worth a close read - no skimming, no tl;rd etc. The question applicable to us - where are the best leverage points in making the D language more successful. > > > Andrei I don't have much influence on the first 4 types of "leverage points" in D, but I have a suggestion for a new "rule of the system" (5th most important type of leverage point). Require reviews from any user before merging their pull requests. There's a number of ways you could implement the requirement, maybe every PR that a user creates needs to have at least 1 review of another PR associated with it. You could require more or less reviews depending on the size of the PR queue. You could also look at developer's "review to pull request" ratio. Just to get an idea, I wrote a script to calculate some of this data (github.com/marler8997/githubstats). Here's the data for dmd, sorted by review to pr ratio: user review/pr reviews open_prs merged_prs closed_prs ZombineDev 25 250 0 7 3 stefan-koch-sociomantic 19.5 39 0 2 0 andralex 17.95833333 431 0 17 7 jacob-carlborg 16.18 809 2 41 7 kubasz 12 12 1 0 0 dkgroot 9 18 0 2 0 trikko 8 8 1 0 0 timotheecour 6.5 65 0 3 7 iain-buclaw-sociomantic 6 6 0 1 0 majiang 6 6 0 1 0 JinShil 5.858895706 955 6 129 28 TurkeyMan 5.529411765 94 1 15 1 thewilsonator 5.111111111 46 2 6 1 Geod24 4.5 117 6 15 5 marler8997 4.155172414 241 0 30 28 dmdw64 4 4 0 1 0 leitimmel 4 4 0 1 0 schveiguy 3.833333333 23 0 5 1 atilaneves 3.727272727 41 1 7 3 DmitryOlshansky 3.166666667 19 1 2 3 tgehr 3.111111111 56 1 16 1 wilzbach 2.946428571 990 25 250 61 FeepingCreature 2.9 29 3 6 1 mathias-lang-sociomantic 2.846153846 111 0 30 9 belm0 2.666666667 8 0 2 1 n8sh 2.5 5 1 1 0 dgileadi 2.5 10 1 2 1 UplinkCoder 2.186813187 199 3 52 36 rikkimax 2 6 1 0 2 EyalIO 2 2 0 1 0 MoritzMaxeiner 2 2 1 0 0 rtbo 2 2 0 1 0 belka-ew 2 2 0 1 0 RazvanN7 1.893333333 284 8 116 26 ntrel 1.846153846 48 2 21 3 nemanja-boric-sociomantic 1.8 9 0 3 2 MetaLang 1.8 9 0 3 2 joakim-noah 1.571428571 11 1 4 2 Darredevil 1.5 3 0 1 1 skl131313 1.5 9 1 3 2 JackStouffer 1.5 3 0 2 0 arBmind 1.5 3 0 1 1 CyberShadow 1.474576271 87 0 53 6 BBasile 1.36 34 0 11 14 Burgos 1.333333333 4 0 3 0 ibuclaw 1.32967033 484 15 293 56 klickverbot 1.327586207 77 0 53 5 kinke 1.186046512 51 1 37 5 quickfur 1.09375 35 0 25 7 don-clugston-sociomantic 1 1 0 0 1 tsbockman 1 5 1 2 2 rainers 0.9832635983 235 5 221 13 WalterBright 0.796728972 1364 15 1607 90 JohanEngelen 0.7575757576 25 0 22 11 John-Colvin 0.75 3 1 2 1 Biotronic 0.75 3 1 1 2 adamdruppe 0.7333333333 11 2 8 5 yshui 0.6666666667 4 0 5 1 mihails-strasuns 0.6086956522 14 0 12 11 mihails-strasuns-sociomantic 0.5 10 0 19 1 Syniurge 0.5 1 0 1 1 aG0aep6G 0.4666666667 7 0 13 2 MartinNowak 0.4323308271 230 8 458 66 LemonBoy 0.431372549 22 7 27 17 LightBender 0.4 2 0 2 3 IgorStepanov 0.3714285714 13 2 22 11 somzzz 0.3684210526 7 0 12 7 yazd 0.3333333333 1 0 3 0 jmdavis 0.3 3 1 6 3 Ingrater 0.2857142857 4 0 12 2 jpf91 0.1904761905 4 0 12 9 nordlow 0.1818181818 2 0 7 4 WalterWaldron 0.1666666667 1 0 6 0 leandro-lucarella-sociomantic 0.15 6 0 31 9 denis-sh 0.1428571429 1 0 6 1 yebblies 0.05793742758 50 5 767 91 braddr 0.05309734513 6 0 102 11 donc 0 0 0 253 17 michelf 0 0 0 4 3 complexmath 0 0 0 5 3 9rnsr 0 0 14 1861 140 Govelius 0 0 0 13 2 wolfwood 0 0 0 2 0 jnschulze 0 0 0 1 0 rawler 0 0 0 0 1 torarin 0 0 0 0 1 mrmonday 0 0 0 4 1 kennytm 0 0 0 20 11 <unknown> 0 0 0 151 71 Poita 0 0 0 4 0 ckamm 0 0 0 3 0 Marenz 0 0 0 0 1 mleise 0 0 0 2 1 dsimcha 0 0 0 3 0 Abscissa 0 0 0 4 1 alexrp 0 0 0 13 10 llucax 0 0 0 15 4 revellian 0 0 0 0 1 timesqueezer 0 0 0 1 0 jcd 0 0 0 0 3 Trass3r 0 0 0 3 5 ohjames 0 0 0 1 0 carlor 0 0 0 3 1 venix1 0 0 0 0 1 wrzoski 0 0 0 0 1 Safety0ff 0 0 0 8 2 bhelyer 0 0 0 1 0 brad-anderson 0 0 0 14 2 p0nce 0 0 0 2 0 NilsBossung 0 0 0 2 2 chadjoan 0 0 0 0 1 ricochet1k 0 0 0 1 1 redstar 0 0 2 37 6 asterite 0 0 0 1 0 dheld 0 0 0 8 0 jmaschme 0 0 0 0 3 biozic 0 0 0 1 0 repeatedly 0 0 0 1 0 edmccard 0 0 0 1 1 jkm 0 0 0 0 1 pszturmaj 0 0 0 1 0 shoo 0 0 0 1 1 ony 0 0 0 1 1 jholewinski 0 0 0 1 0 bioinfornatics 0 0 0 1 0 monarchdodra 0 0 0 3 0 dsagal 0 0 0 1 0 jkrempus 0 0 0 1 1 sgraf812 0 0 0 1 0 glycerine 0 0 0 0 3 eskimor 0 0 0 2 1 lionello 0 0 0 15 10 hpohl 0 0 1 51 22 jordisayol 0 0 0 0 1 huhlig 0 0 0 1 0 AlexeyProkhin 0 0 0 1 1 Numpsy 0 0 0 1 0 atlant2011 0 0 0 1 0 Kapps 0 0 0 2 0 elvisxzhou 0 0 0 1 0 WebDrake 0 0 0 1 1 Dgame 0 0 0 0 3 AndrewEdwards 0 0 0 7 2 D8174 0 0 0 1 0 schuetzm 0 0 0 5 4 nazriel 0 0 0 1 0 yglukhov 0 0 0 2 2 Yoplitein 0 0 0 1 0 Orvid 0 0 0 9 3 andron 0 0 0 6 1 SSPkrolik 0 0 0 1 0 MasonMcGill 0 0 0 0 1 qchikara 0 0 0 4 0 Hackerpilot 0 0 0 2 1 damianday 0 0 0 0 1 dcarp 0 0 0 1 0 jasonbking 0 0 0 3 0 jcrapuchettes 0 0 0 1 0 teufelsmangagirl 0 0 0 0 2 ltcmelo 0 0 0 1 0 dragoon2014 0 0 0 1 0 jamestn 0 0 0 1 0 9il 0 0 0 4 3 etcimon 0 0 0 3 1 tramker 0 0 0 7 4 asuhan 0 0 0 0 1 Temtaime 0 0 0 1 2 gchatelet 0 0 0 5 2 deadalnix 0 0 0 1 1 cqexbesd 0 0 0 2 0 smolt 0 0 0 6 1 legrosbuffle 0 0 0 4 1 davispuh 0 0 0 1 1 Kozzi11 0 0 1 6 0 burner 0 0 0 1 1 Rajeep 0 0 0 0 1 dsp 0 0 0 1 2 luismarques 0 0 0 6 1 andrej-mitrovic-sociomantic 0 0 0 1 0 AndrejMitrovic 0 0 1 7 3 Cauterite 0 0 0 3 5 TheDharc 0 0 0 0 1 flaviommedeiros 0 0 0 1 1 TungstenHeart 0 0 1 0 0 blm768 0 0 1 0 0 landaire 0 0 0 2 0 Computermatronic 0 0 0 0 1 joakim-brannstrom 0 0 0 1 0 alphaKAI 0 0 0 1 0 rjmcguire 0 0 1 0 0 calexHG 0 0 0 2 0 Passw 0 0 0 3 0 Pursche 0 0 0 1 0 stevepeak 0 0 0 1 0 Jebbs 0 0 0 1 0 NVolcz 0 0 0 1 0 ahmetsait 0 0 0 0 1 andrey-zherikov 0 0 0 1 0 GooberMan 0 0 0 1 0 dhasenan 0 0 0 0 2 mathias-baumann-sociomantic 0 0 0 1 1 MaskRay 0 0 0 1 0 epi 0 0 0 1 0 e-y-e 0 0 0 1 0 e10s 0 0 0 2 1 dukc 0 0 0 0 1 ka7 0 0 0 2 0 DrInfiniteExplorer 0 0 1 0 0 s-ludwig 0 3 0 0 0 nrTQgc 0 0 0 1 0 zachthemystic 0 0 0 1 0 MikeWey 0 2 0 0 0 ReneZwanenburg 0 2 0 0 0 veelo 0 2 0 0 0 jll63 0 0 0 1 0 nmtigor 0 0 0 1 0 dunkyp 0 0 0 2 0 jercaianu 0 0 0 1 0 carun 0 2 0 0 0 monkeywithacupcake 0 0 0 0 1 bausshf 0 1 0 0 0 gautam-kotian-sociomantic 0 1 0 0 0 lgvz 0 0 0 1 0 kubo39 0 0 2 3 1 GabyForceQ 0 0 0 1 0 aliak00 0 1 0 0 0 gapdan 0 0 0 1 0 | |||
August 26, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Jonathan Marler | On Friday, 24 August 2018 at 03:06:40 UTC, Jonathan Marler wrote: > I don't have much influence on the first 4 types of "leverage points" in D, but I have a suggestion for a new "rule of the system" (5th most important type of leverage point). Require reviews from any user before merging their pull requests. There's a number of ways you could implement the requirement, maybe every PR that a user creates needs to have at least 1 review of another PR associated with it. You could require more or less reviews depending on the size of the PR queue. You could also look at developer's "review to pull request" ratio. Interesting idea. > Just to get an idea, I wrote a script to calculate some of this data (github.com/marler8997/githubstats). Here's the data for dmd, sorted by review to pr ratio: […] Interesting data as well. Seeing that relatively few have a review/pr ratio > 1, you may be onto something. (The list seems to have an issue with ordering though, for those that reviewed without having PRs. Attributing them a ratio of > 1 would be fairer than 0). | |||
August 30, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
 Posted in reply to Laeeth Isharc | On Monday, 20 August 2018 at 12:26:25 UTC, Laeeth Isharc wrote: > On Monday, 20 August 2018 at 11:55:33 UTC, Joakim wrote: >>>> "So how do you change paradigms? Thomas Kuhn, who wrote the seminal book about the great paradigm shifts of science, has a lot to say about that. In a nutshell, you keep pointing at the anomalies and failures in the old paradigm, you keep coming yourself, and loudly and with assurance from the new one, you insert people with the new paradigm in places of public visibility and power. You don’t waste time with reactionaries; rather you work with active change agents and with the vast middle ground of people who are open-minded." > > (Quoting from the article I think). > > Kuhn and Lakatos. Paradigm shifts don't take place when the dominant paradigm is defeated by logical or empirical means. Paradigm shifts take place when for some reason people say "how about we stop talking about that, and start talking about this instead". Not sure why you'd call that anything other than defeat. :) > I think he described certain political changes in the Western World beginning in the mid to late 60s rather well. I don't think it describes how changes in the sphere of voluntary (non-political ie market and genuine civil society) activity unfold. Supposing it were a good idea (which it isn't), how would one be able to to insert people in places of public visibility and power who put forward a point of view that is very different from the prevailing one? Only via a program of entryism, and I don't think that in the end much good will come of that. By convincing those with power/visibility that the contrary view is worth integrating? Look at Microsoft's about-face on open source over a couple decades, going from denigrating it to buying open-source producing or supporting companies like Xamarin and Github and open-sourcing several of their own projects, as an example. > So I think the original author has cause and effect the wrong way around (not too surprisingly because he is talking about things that relate to politics and activism). [NB one shouldn't mention the Club of Rome without mentioning what a failure their work was, and it was predictably and indeed predicted to be a failure for the exact same reasons it failed]. > > It isn't that you insert people representing the new paradigm in positions of influence and power. > > It is that people from the emerging new paradigm - which is nothing, a bunch of no-hopers, misfits and losers viewed from a conventional perspective - by virtue of the fact that it has something useful to say and has drawn highly talented people who recognise that start ever so slowly to begin things and eventually to accomplish things - still on the fringes - and over time this snowballs. After a while turns out that they are no longer on the fringes but right at the centre of things, in part because the centre has moved. > > The best illustration of this phenomenon was I think in a work of fiction - Neal Stephenson's Cryptonomicon. I never expected someone to write a novel based on a mailing list - the cypherpunks. It was about as surprising to me then as it would be to see Dlang - the movie - today. And of course that itself was an early indication that the ideas and ways of thinking represented by what was originally quite a small community were on the ascent. I agree that she's looking at it from the point of view of governmental change for her environmental agenda, whereas the market is more likely to have entirely new institutions- it used be new _companies_, but with the internet it's now increasily decentralized operations like the community behind bitcoin or bit torrent... or D- form that become much more important than the old ones: creative destruction. So, significantly open-source Android replaces mostly closed Windows as the dominant OS used by most consumers for personal computing, rather than Microsoft really getting the new religion much. >>>> This pretty much reflects what Laeeth always says about finding principals who can make their own decisions about using D. "Places of public visibility and power" for D are commercial or open-source projects that attract attention for being well done or at least popular. > > Well - I understand what you mean, but I don't recognise this as being my point. Principals who can make their own decisions probably aren't today highly visible and visibly powerful. The latter comes much later on in the development of a project, movement or scene and if you're visible it's a tax that takes time away from doing real work. By the time you're on the front cover of Time or The Economist, it's as often as not the beginning of the end - at least for anything vital. You're misreading what she wrote: she only said that you place new people in positions where they have some visibility or power, again because of her emphasis on government change, not that you convince the already _highly_ visible/powerful to go your way. Since the former can be almost anything depending on the context, I gave examples of who that might be for D. >> We're doing both: most of the material on the D blog and my own D interviews are not with corporate representatives. We could stand for more of the latter though, especially the big successes, because people are more influenced by them. > > I'm not saying it's a bad thing to go for big stories. But it's a mistake to place the attention people today naturally tend to. It doesn't matter what influences most people - it matters what influences the person who is poised on the edge of adopting D more widely, adopting D as a beginning, or would be if they knew of the language. The latter is quite a different sort, I think. To know about D in the first place, you might have the sort or programmer who's not going to actually use it become aware of it and mention it to the guy who would use it but never heard of it. In other words, there are different ways of getting to the people who would use D: I think this is one of the better ones, but we need to use several ways. One of the best is a track record of successful projects using the language. > Liran at Weka picked up D because he saw Kent Beck post on Twitter about Facebook's Warp written in D (or maybe it was a linter) and it seemed like an answer to a particular problem he had (if I am remembering correctly). It wasn't because of a grand thing - it was because of a little thing that seemed like it might be a creative solution to a real problem. > > Signal:noise is much higher away from the limelight too. By far better to have a high share of attention in some specific domains or interest groups than to have a low share of attention of some enormous market. Sure, we need to try several ways to attract attention. >> Many devs use large corporate deployments as a litmus test of whether they should explore a new tech. To the extent that we've never published a blog post about Weka, only offhand mentions like when Andrei visited Israel, that is a big marketing failure for D. >> >> I know the Weka guys are very busy, but the further success of D will only help them too, so they're undercutting themselves by not making sure that blog post gets done. > > Well, someone could just take the key insights and experiences from their talks, write them up, check with them and post. The latter are a considerable commitment already for a startup that's hitting a revenue growth phase. There are lots of things for them to be busy with beyond just the technology. Yeah, that could work. >>>> Finally, regarding leverage, I keep pointing out that mobile has seen a resurgence of AoT-compiled native languages, but nobody seems to be trying D out in that fertile terrain, other than me. >>> >>> I did try, but it's not exactly easy to make a complete app in D, even on Android. It would be great if there were some way to automatically wrap the APIs. >> >> Right now, the Android port is more suited for writing some performant libraries that run as part of an existing Android app. The kind of polish you're looking for will only come with early adopters pitching in to smooth out those rough edges. > > If we had autowrap for JNI and could dump the types and method prototypes as part of the pre-build process, what would the next stage be to be able to just call Android APIs from D and have them work? JNI isn't that bad (I know it's deprecated) and I used it already from D in a semi-wrapped way. So I wonder how much more work it would be to have autowrap for JNI. > I didn't use reflection on the Java side because I wasn't wrapping that much code. Are there XML descriptions of Android APIs you could use to generate wrappers? No idea, not something I've looked at nor plan to, as I mentioned to you in over email recently. I'd like to use mostly D with not much Java on Android, so I'm not too worried about the interface right now. | |||
August 31, 2018 Re: [OT] Leverage Points | ||||
|---|---|---|---|---|
| ||||
Posted in reply to Joakim | On Thursday, 30 August 2018 at 11:45:00 UTC, Joakim wrote:
>> (Quoting from the article I think).
>>
>> Kuhn and Lakatos. Paradigm shifts don't take place when the dominant paradigm is defeated by logical or empirical means. Paradigm shifts take place when for some reason people say "how about we stop talking about that, and start talking about this instead".
>
> Not sure why you'd call that anything other than defeat. :)
FWIW, it's the point of Lakatos's work: he argues that a paradigm can't be defeated by logical or empirical means. It takes zero effort to not do anything, so status quo is easily maintained.
| |||
Copyright © 1999-2021 by the D Language Foundation