Hacker News Reader

Hacker News

Comment by jaems33 | Hacker News Reader

Parent Full thread

jaems33

Quite a bit. Looking at GIMP's roadmap:

- Improvements in the text handling

- Automatic layer boundary management

- Filter layers and "Layer effects"

- Non-destructive editing

- Auto-anchoring of floating selection - or better, get rid of floating selections!

- Script recording and playback

- "Smart objects"

Even for me as a mere moderate power-user (I design websites/apps and do photo adjustments), there's four things on that list that are absolutely essential. And I'm willing to pay the extra dollars for it given how it enhances the final product.

View on HN

That's quite a list. I wonder why it takes so long for these features to be supported. Perhaps what's necessary is a new paradigm of development: a core application with composable modules developed by a community on the basis of need. Evolution of standards will be tricky to manage but it seems likely to be inevitable.

It takes so long because the GIMP developers have a long-standing tradition of telling people with requests to go jump off a cliff if the feature being requested isn't directly useful to the developers. They have a fairly typical failing in assuming that because the source is available, anyone asking for features has the ability and the time to implement those features themselves, when often that is not the case.

If they're not being paid to work on it, why would you expect them to prioritize things they're not interested in working on or not going to use?

Because they want people to use their software?

Of course, if you don't care whether or not anyone uses your software, telling people with questions and requests to jump off a cliff is entirely reasonable.

I don't expect them to. I'm just answering the question as to why features take so long to get implemented.

So I'm thinking one way to reduce the barrier for entry into participation is to have the bulk of the application's features in a higher level language (rather than C), with the core in a high performance language. Unfortunately this is not something I'd have the time to pursue.

Reply on news.ycombinator.com