I really didn't intend it as a swipe against Miguel, but I think it does inform his perspective. And I meant "platform fan" in the advocacy sense: he has a tendency to "fall in love" with favorite products. That's not uncommon in the general population (it's pretty much the norm at HN!), but it certainly is among core Linux people who tend to prefer doing new things in different ways.
> His early world was Midnight Commander, Gnome, Gnumeric, and Evolution.
Clones of Norton Commander, Windows, Excel, and Outlook. To be fair, Gnome 1 wasn't really a "clone" (though it did mimick more than innovate) and mc was chasing a Symantec product, not a Microsoft one.
But to claim that these were innovative new projects is silly. Miguel's career has been one of seeing something he loves in an existing product and duplicating it in his preferred free software environment. There's no shame there. But it's absolutely the same thinking that drove the Mono project.
Pedantically speaking, I wrote Gnumeric not because I loved Excel, or enjoyed using it or because I was trying to be innovative. In fact, I did not even know how to use a spreadsheet at the time.
Gnumeric was the product of the mood in the early days of Gnome: we need to provide this to have a complete desktop offering and we were talented hackers that could get it done. So I did.
Once I started, I enjoyed writing a spreadsheet every second of it. It was both a very educational process, but also the one that made me grow fonder of strict compiler warnings and errors. And from this experience, most Gnome software after this was compiled with warn-as-error.
>And I meant "platform fan" in the advocacy sense: he has a tendency to "fall in love" with favorite products.
Well, for one, C#/CLR was not a favorite product when he fell in love with them, were they? At the time lots of people thought MS was foolish to try to overtake Java, and it took a couple of years for people to accept and like .NET.
>But to claim that these were innovative new projects is silly.
Sure, they weren't, but it's not like Linux in general has given us much innovative things in the desktop space.
Workspaces? You mean virtual desktops? Those date back 1986, implemented at Xerox PARC (and patented!).
As for appstores, I assume you're talking about the packaging systems, which are similar only if viewed from 50,000 feet.
The packaging systems were not stores. You couldn't buy packages. They didn't serve as an agent to sell other people's personally submitted packages. They had "apps" but no "store". The applications weren't sandboxed to make it relatively safe for them to sell 3rd-party submitted packages. The applications were expected to be open-source.
The only similarity to the app stores is that they had a repository of software and an automated installation system used to install it.
Plus the dependencies. With app stores you (usually) get one program and that's it. It works. With package systems you have to install several dependencies for each, sometimes an absurd number (say, 100 packages to get Gnome). And then there's dependency problems (had my share in Debian at older times).
At the core, dependencies are a good thing, although there's downsides to that as well.
I have no experience with desktop app stores, but you wouldn't get Gnome and similar large and complex software from one, would you?
Which makes your comparison pretty unfair.
You can't name innovative things amongst a crowd that rejects that innovation can happen. They'll tell you someone else did it earlier, even when they didn't.
For instance, cars were not an innovation according to hacker news, because Horses!
> His early world was Midnight Commander, Gnome, Gnumeric, and Evolution.
Clones of Norton Commander, Windows, Excel, and Outlook. To be fair, Gnome 1 wasn't really a "clone" (though it did mimick more than innovate) and mc was chasing a Symantec product, not a Microsoft one.
But to claim that these were innovative new projects is silly. Miguel's career has been one of seeing something he loves in an existing product and duplicating it in his preferred free software environment. There's no shame there. But it's absolutely the same thinking that drove the Mono project.