In two blog entries [1] [2], LionKimbro talks about how (my interpretation) it's hard to get support for one's software projects.

Lion, i feel your pain.

I think that agreeing to meet everyone's needs is too much committment for me. But i'm not sure that's really the problem. What bugs me the most as a programmer is not lack of support needs, but lack of any feedback whatsoever. If people said they liked the project, that would be nice. If people said they thought it was mostly a waste of time and why, that would be very useful. If people said they don't really understand the project, that would be very useful. Even if they said aloud what I interpret silence to mean ("the project's okay but i'm not interested enough to work on it or use it"), that would be better than nothing.

So I think we just have to make more of an effort to solicit and to give opinions on each other's projects. Hence this page.

If you'd like to hear other's opinions on your projects, please create a section heading for yourself. List your projects, and link to descriptions of them. Please note that you may receive harsh criticism so if you don't want to hear that, don't list your projects. Also, list your target audience.

Everybody else, when you see a project listed here, and you are in the target audience, for the love of wiki don't be silent! We would rather hear "I think this project is a dumb idea", or "I don't have time to understand this project" than hear nothing at all (if you can say why its dumb, that's even better). If we wanted silence rather than criticism, we wouldn't list here.

Also, if you don't have time to read about the project right now, or to write a long answer, fine. Better a short answer than silence. If no one groks the project, that's useful info in itself. You can change your answer later if you want to

If you're not in the target audience, feel free to be silent.

Some questions that you might like to answer:


Bayle Shanks. target audience: programmers who do wiki-related stuff

The projects that I would currently like feedback on are:

WikiGateway – I think I know what the goal of the project is, and I think it would be useful. I'm only familiar with my own code, however, and I haven't installed the gateway, and thus it was easier to work on my own implementation of WebDAV instead of relying on the wiki gateway. Doing it myself seemed more rewarding in terms of my learning experience, since there was only one thing to learn, and not two, and because I knew the infrastructure intimately. The knowledge I gained about WebDAV remains useful for future projects of mine, whereas knowledge I gained about the wiki gateway would have a shorter halflife: There's risk that it will be abandoned, unpopular, etc. Given that I have finite resources, that explains why I didn't join in. As for publicity: I knew a bout it. I don't know if JanneJalkanen knew about it when he wrote about WebDAV and wikis, for example.

CommunityProgrammableWiki – I think I know what this is about, and I don't think we need it. I rarely felt the need to actually modifiy wikis that were not my own. Basically it happened once or twice with Meatball. In these cases, the social barrier was much higher: Convincing Sunir that something had to be done, convincing the community, volunteering to do it, implement it, test it, maintain it. I don't think it would be different for other projects: Modifying the code needs intimate trust, and thus a newbie would not be allowed to change the way the software works in significant ways. Localized bug fixes, yes, paradigm shifts, coding style, modularization, no. Since those people are few, they can get write access to a shared version control repository and servers anyway. There very rarely is a real need that a community programmable wiki would solve.

PICA – I think I know what this is about. I think something like the microformats wiki/blog shows that something like it might work for wikis. I don't think it will tremendously boost the output of PICA members, however. The kind of developers that would come to the PICA site and use something on it is self-selected. These kind of people would have found similar ways of collaborating with other developers, too. So we're facilitating something that isn't too hard to do, anyway.

CommunityProgrammableWiki – I think part of the idea is to reverse the whole "only a few have access & care about the code" thing, sort of like how wiki reversed the whole "the system administrator document maintainer is the one who manipulates the pages" thing.

That is: Turn the maintenance into a social experience, and teach people how to participate in the maintenance.

OddMuse has a bunch of formatting rules, and we have all these various ways of teaching them to people.

Perhaps it can work with code, as well.

When MattisManzel asks for a feature, you can say, "read up in this area." Then he goes and asks questions about what he sees. We point him to the right areas. Some things will be too hard, some things won't be too hard. Once he gets accustomed to the layout, and the language, perhaps, with time, he can work on more complicated things.

We can also reshape the "language of functions" to make the language more accessible to him, as well, more dynamic.

So I can see a variety of people at different levels of understanding making ways for each other to manipulate the code in a social experience.

So the root assumptions- "no one would have the expertise or care enough to bother-" are what we believe we may be able to subvert.

PICA – The goal of PICA is not to boost our output, here in the CommunityWiki. The goal of PICA is to boost the output of collaboration software developers in general. EugeneEricKim of BlueOxen made a point that what is most needed right now is for software interoperability.

We're not talking just wiki, we're talking about the whole 9 yards of pretty much everything that isn't accounted for by an existing body. That is, everything ProfessionalAmateur? that is not a microformat!

This is precisely the point: Those developers have no place for finding others at this time. I mean, if you want to hook just two apps together, then you have two people to talk with. But the idea is that general interfaces are made for general use, which requires general knowledge of others.

The strategy is to:

  1. Make a place that collects specifications by scanning, then posts to a blog.
  2. As readers become accustomed to it, focus more on the wiki end of things.

Actually, we intend to use OddMuse with the FrontPage thing. That's what we want to start with.

WikiGateway is a good thing, I said on WikiSym that I would support it and that it should be supported. It just took some time for me to come in the situation where I really need it. I think it is ahead of its time, maybe two years early. You have just to be a bit patient. I will now implement a ProWiki driver and contribute.

CommunityProgrammableWiki is imho more a spike to show what can be done in a wiki. ProWiki could easily implement it, but it seems not worth the effort, at least currently. There are too many security risks. Standard ways of OS code contributions seem sufficient.

I don't understand the PICA idea and I don't care much about standards. Until I need them someday I try not to get my feet wet.

Thank you Alex! That's very useful information for me.

Perhaps we should move further discussion resulting from the replies to the relevant pages? That will leave this space free for just getting the initial feedback.

Thanks Helmut!

  • WikiGateway: Ok, a GET and POST library that groks different wiki responses… Could this be written in JavaScript to run from the browser without installation…

For markup standardization I vote for PlainText and PlainLink.

User identification, qualification and weight must be determined.
Code is not inserted unless enough workers with enough weight vote it in.
Weight might be earned through work which is not 'voted-down' or revereted.
Weight might be purchased with real money used toward the cost of hardware and hosting.
Probably need some sort of false-bottom such as "two signatures required".
Automatic Determination of possibly malicious code … what's that EmacsLisp? package that checks spreadsheet entries…
Qualified workers should not be stopped by an oppressive owner.

Thanks Patrick! Great ideas. I am already planning to do some of them.

btw if you ever feel like it, lemme know if you want to work on CommunityProgrammableWiki.

Define external redirect: ProfessionalAmateur FreeHardware EmacsLisp

EditNearLinks: JanneJalkanen EugeneEricKim ProWiki JavaScript