Development quote of the week [LWN.net]

The hard economic truth for engineers is that technical excellence is overwhelmingly irrelevant, and the history of our profession is full of cases that demonstrate this. Javascript as a language, for instance, largely came to dominate the browser ecosystem only due to the Netscape marketing department, who decided on a clever marketing ploy to circumvent Sun Microsystems and give their own new language cachet by hijacking existing name of Java. While we can’t attribute all of Javascript’s success to this naming, it is quite likely this trick was sufficient to convince enough non-technical stakeholders to make a significant difference in early adoption. These kinds of marketing tricks and positioning are likely the most fruitful path to reaching a large technical audience.

— Stephen Diehl (Thanks to Paul Wise)

to post comments

Development quote of the week

Posted Jun 4, 2020 6:31 UTC (Thu) by roc (subscriber, #30627) [Link] (6 responses)

That blog post raises an obvious question: if marketing/enterprise-ish issues are so dominant, correctness doesn't matter, etc ... how has Rust been so successful?

The answer is not "because Mozilla poured money into it". Mozilla never invested hugely in Rust and that investment has been declining for a long time now. Having said that, Mozilla's backing did help a lot.

Development quote of the week

Posted Jun 4, 2020 6:52 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link]

90-s were largely the time of software adolescence. Nobody particularly cared about reliability or long-term maintainability. Also not many companies listened to actual engineers and it was much easier to sell buzzword-laden crap.

I fondly remember how every largeish corporation offered an EJB server, with stuff like BEA WebLogic selling licenses costing many thousand dollars per CPU.

Since then the software world has grown up a bit. I get really happy and fuzzy feeling from all the work done on reproducible builds, proven correctness, fuzzers, etc. None of this was even conceivable in 90-s (except maybe in some niche domains).

Development quote of the week

Posted Jun 4, 2020 7:02 UTC (Thu) by synacktiv (subscriber, #86420) [Link] (3 responses)

I am not sure that Rust is "so successful", and it looks to me that Go is wildly more "successful", while being less correct than Java. Same thing for Javascript, which did not seem to turn people into true believers at the same rate Rust does, yet achieved dominance in its ecosystem.

I believe language adoption is only distantly driven by the technical merits of this language. As for Rust, I *believe* that, contrary to Go, people like it for its technical merits, but nobody would have heard of it if it wasn't for Mozilla. As for the name effect, Javascript was available before Flash existed, and for a while Flash was a lot more popular.

Development quote of the week

Posted Jun 4, 2020 16:06 UTC (Thu) by khim (subscriber, #9252) [Link] (2 responses)

Go is definitely more successful, but remember the context of the original quote: we are talking about Haskell here.

Let's go back to 2014: Rust is not in Top50 yet, but Go and Haskell both are in Top50, Haskell is #49, Go is #46. F# (not quite Haskell, but kind of it's bastard cousin) is in Top20.

Five years ago, 2015th: Go went back below Top50 level, Rust is still there, Haskell is #48, but F# is no longer in Top20, it's #39 instead.

Four years ago, 2016th: Go is still out of Top50, but Rust in #44 while F# and Haskell are higher: Haskell is #38 and F# is #42.

Three years ago, 2017th: Go went from #55 to #10! In one year! That's quite a success… and shows that all these ratings are quite volatile… but I digress. F# and Haskell are best buddies now: #40 and #41. And Rust is near them, too. Just above: #38.

Two years ago, 2018th. Go is still in Top20, F#, Haskell and Rust are all in Top 50: F# is #36, Haskell is #44 and Rust is between them, #39.

One year ago, 2019th. Go is steady in Top20, F#, Haskell and Rust are in the same position as in 2018th: F# is #30, Haskell is #42, Rust is #38.

Today, 2020th. Both Go and Rust are in Tip20... while F# left Top50 and Haskell… still at rank #46.

So yes, Rust is quite successful. Not as successful as Go, but definitely better than Haskell.

The most interesting is fate of F#. It's rise and fall should teach us something, I think. I'm just not entirely sure what…

Development quote of the week

Posted Jun 5, 2020 11:59 UTC (Fri) by tao (subscriber, #17563) [Link]

I suspect that Go's jump in popularity might be connected to Kubernetes.

Development quote of the week

Posted Jun 9, 2020 2:07 UTC (Tue) by ncm (guest, #165) [Link]

Tiobe is not just volatile. It is spurious. Anybody supporting an argument based on Tiobe identifies their argument as equally spurious.

If you could rally an argument based on some other, unavoidably better source, you would.

In fact Rust has achieved a kind of success, in that a noticeable amount of defensible code has been written in it, and some thought leaders make a show of coding it. However, the total number of programmers using Rust in production is probably less than the number of programmers who switch to C++ in any given week. In ten years, Rust might reasonably be counted a success. At this stage it is can, at best, be noted as serious and interesting.

Development quote of the week

Posted Jun 4, 2020 11:56 UTC (Thu) by rahulsundaram (subscriber, #21946) [Link]

> The answer is not "because Mozilla poured money into it"

Mozilla using Rust in Firefox was a very important step in getting Rust into all the Linux distributions paving the way for other uses like librsvg etc. It's not just about the money. It's more the visibility that Mozilla provided that was crucial initially

Development quote of the week

Posted Jun 4, 2020 8:52 UTC (Thu) by mti (subscriber, #5390) [Link] (1 responses)

I don't think it was marketing that made Javascript big but rather ease of use and availability. In the beginning it was very easy to add nice features to a webpage by adding some Javascript. Things like validation of forms. And if you saw some nice feature someone else had done you could just look at the source and copy the code. Even non-programmers could copy a snippet and add it to their webpage.

Java and Flash were harder to use. And Java applets had irritating long startup times.

Development quote of the week

Posted Jun 19, 2020 3:04 UTC (Fri) by ghane (guest, #1805) [Link]

> And if you saw some nice feature someone else had done you could just look at the source and copy the code. Even non-programmers could copy a snippet and add it to their webpage.

This. Very much this.

(Speaking as a non-developer CTO in those days)

--
Sanjeev

Development quote of the week

Posted Jun 4, 2020 13:32 UTC (Thu) by cesarb (subscriber, #6266) [Link]

The success of Javascript has nothing to do with marketing, and everything to do with the fact that it was the only alternative. If you wanted to run client-side scripts on the dominant browser everyone used (Netscape Navigator), it was Javascript or nothing. You could also use plugins (like Flash or Java), but as far as I recall, they appeared as their own separate element on the page (not to mention being much heavier to use, having to be compiled and distributed separately, instead of just writing a couple of inline tags or attributes).

Once every site started to use and even depend on Javascript, any new browser had to also implement Javascript, cementing its dominance.