4 ms·

So far, it looks like nobody is commenting on Dave Winer’s own work. Everybody has been influenced by Engelbart, but nobody uses NLS. Merely the ideas went int

by Decade 11y ago

So far, it looks like nobody is commenting on Dave Winer’s own work. Everybody has been influenced by Engelbart, but nobody uses NLS. Merely the ideas went into all the computers that we use today. Engelbart was so frustrated because he had more great ideas, but they were difficult to develop and he couldn’t get funding. Winer specifically mentions Userland Frontier. It was a useful tool, which I’m sure helped with the development of RSS, just as NeXT Project Builder helped with the development of WWW. The problem with the Engelbart comparison is that there are countless other development environments, many of them free and/or not in a dialect of C and/or available via git, while the Mother of All Demos was unique. I think the open-sourcing of Frontier was too little, too late. Frontier is an impressive achievement, but I don’t see why it should be interesting to me.

dotandimet 11y ago
I've been following Dave for a long while, and I find he excels at leading you towards interesting ideas, starting conversations, discovering and inspiring interesting work by others, experimenting with cool stuff, occasionally just inventing influential technologies. Blogging, RSS, MP3 sharing, podcasting, HTML UIs, sites as apps, REST APIs/Web RPC... I used Frontier in 1998 as a static site generator, and later as an RSS reader and a blogging tool. Frontier was a very interesting technology, and it was probably part of what made Dave so productive. Imagine a single program that is a web server and a database, a system scripting tool, a scheduler (like cron or Jenkins), an editor and an outliner and a scripting language runtime and an IDE... That's what Frontier was, and building each part of it leveraged the rest, accreting functionality to itself as it grew like coral. Frontier's integrated approach went against the prevailing attitude in software development, that of UNIX, Open Source and the Web, of worse-is-better, loosely-coupled, modular and specialized tools. But by embracing the web's open protocols and loosely-specified text-based APIs, Dave got more than a decade of remarkably productive extra mileage out of it.

by Decade 11y ago

dotandimet 11y ago

I've been following Dave for a long while, and I find he excels at leading you towards interesting ideas, starting conversations, discovering and inspiring interesting work by others, experimenting with cool stuff, occasionally just inventing influential technologies. Blogging, RSS, MP3 sharing, podcasting, HTML UIs, sites as apps, REST APIs/Web RPC... I used Frontier in 1998 as a static site generator, and later as an RSS reader and a blogging tool. Frontier was a very interesting technology, and it was probably part of what made Dave so productive. Imagine a single program that is a web server and a database, a system scripting tool, a scheduler (like cron or Jenkins), an editor and an outliner and a scripting language runtime and an IDE... That's what Frontier was, and building each part of it leveraged the rest, accreting functionality to itself as it grew like coral. Frontier's integrated approach went against the prevailing attitude in software development, that of UNIX, Open Source and the Web, of worse-is-better, loosely-coupled, modular and specialized tools. But by embracing the web's open protocols and loosely-specified text-based APIs, Dave got more than a decade of remarkably productive extra mileage out of it.