Path: csiph.com!weretis.net!feeder6.news.weretis.net!news.misty.com!news.iecc.com!.POSTED.news.iecc.com!nerds-end From: gah4 Newsgroups: comp.compilers Subject: Re: Are there "compiler generators"? Date: Tue, 31 May 2022 16:55:22 -0700 (PDT) Organization: Compilers Central Lines: 30 Sender: news@iecc.com Approved: comp.compilers@iecc.com Message-ID: <22-05-066@comp.compilers> References: <22-05-054@comp.compilers> <22-05-058@comp.compilers> <22-05-063@comp.compilers> <22-05-065@comp.compilers> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8bit Injection-Info: gal.iecc.com; posting-host="news.iecc.com:2001:470:1f07:1126:0:676f:7373:6970"; logging-data="67424"; mail-complaints-to="abuse@iecc.com" Keywords: design, comment Posted-Date: 31 May 2022 22:21:19 EDT X-submission-address: compilers@iecc.com X-moderator-address: compilers-request@iecc.com X-FAQ-and-archives: http://compilers.iecc.com In-Reply-To: <22-05-065@comp.compilers> Xref: csiph.com comp.compilers:3038 On Tuesday, May 31, 2022 at 8:20:08 AM UTC-7, Hans-Peter Diettrich wrote: > On 5/30/22 2:53 PM, Hans-Peter Diettrich wrote: > > analysis or translation of computer languages to be on topic. -John] > What are those "computer" languages? I'd prefer "formal" languages > (Chomsky...) instead. E.g. Meta§ also was used for DNA analysis. > [I was going to say artificial languages but I don't think we have anything > useful to say about Esperanto. In practice it hasn't been very hard to keep > discussions more or less on topic. -John] Definitely interpreters and macro processors have been discussed, though some might not call them compilers. Also text processors like TeX. I am wondering, though, about (human) language translators. It seems that many use non-deterministic AI systems, and so are fundamentally different from most of what is discussed here. If you parse Esperanto with a Flex/Bison parser then it should be fine here. If you parse Fortran with a deep neural net, then maybe not. [That's essentially what I've been thinking. In the 1950s and 1960s there was a lot of work trying to do human language translation using formal methods and it worked very poorly, sort of adequate for translating technical manuals, not for anything else. The breakthrough was when someone at Google realized that the Candian parliament's Hansard, the transcript of debates, had high quality parallel French/English translations going back a century and they fed it into their machine learning systems. -John]