FA+
Shop
Ferrox: status update
17 years ago
The veil has been lifted! I'm going to make periodic (for some value of "periodic") reports of Ferrox progress from here on out. This is about the FA equivalent of painting a gigantic target on my chest, I'm sure. Here's what the hell is going on.
First of all, for the 98% of people who still don't know, Ferrox is a rewrite of FA, from the ground up, in Python. It also has a bit of a rocky delay-filled history. It's been in some form of planning (read: "we should do this sometime") since at least last March, but for whatever reason there wasn't really a call for coders until August, and the list wasn't finalized until I guess sometime in September. Initial setup and speccing happened on and off from there on out, and the first real code commit didn't happen until early November. It's been slightly rough going from there on, as we all had to get familiar with the combination of language/libraries/platform we'd ended up with, so small changes often turned into an entire evening of poring over a manual. We also went through at least four different image-handling libraries and discovered they all sucked, at which point we had to find something else and rewrite the existing code again. That's all more or less past now and things are speeding up a bit.
Since this doesn't seem to be actually mentioned anywhere, the current list of coders is:
thecrypto (illustrious leader)
indicoyote
latiass
lexyeevee
kalu
net-cat
verix
Due to a variety of extenuating concerns, not all of these people are actively contributing at the moment. Who is and who isn't is strictly classified (and rude!), but I will say that I am, and I wish more people were. Some of us just have school/work consuming time like Gabe Newell on donuts.
(If I may soapbox for a moment: honestly, it rather sucks being expected to build a fairly large, vaguely backwards-compatible system that will make users of the old crumbling system happy with tools I have never used before.. and have that userbase constantly make Duke Nukem Forever jokes to my face. Thanks, furs, very inspiring.)
Anyway, current status of Ferrox right now:
- Registration and login and all that boring crap works, complete with an actual damn captcha, for all the good the stupid things do.
- Permissions system, well, exists. I guess there's not a lot to be said there.
- Admin panel is started. I'm basically making admin things up as I think of them, since I am not an admin and have no idea what current FA can and can't do.
- Gallery has basic functionality. Upload, look at stuff. Tagging/searching is roughly working.
- News and journals exist, can post/edit/view; no comment support yet.
- Clone of the current theme exists. Support for different themes and color schemes exists. There's also an alternate theme I made up, but I think I was trying to keep too much in the spirit of current FA and made it too bland.
- I'm doing notes right now; they're almost done.
- I'm trying to aggressively refactor and clean up as I go, so the code stays as compact and simple as possible. Always an uphill battle, but seems to be going okay.
3300 lines of code, 1100 lines of template, seventy or so commits.
Not much in the way of user-facing stuff (watches, profile, whatever) has been done, but a lot of that stuff is really just icing that can be banged out in a few hours. What I mostly want is a fast, flexible, and robust search backend we can use everywhere. Ah, I can dream.
Feel free to watch me; I will try to post something every week or so, and possibly throw up screenshots sometime. In theory, this will get my ass in gear so I actually have something to say every week. 8)
I might also brainstorm here if I don't know how to approach some feature; I'm not an artist, after all, so I am occasionally guessing wildly at how people might use this thing.
First of all, for the 98% of people who still don't know, Ferrox is a rewrite of FA, from the ground up, in Python. It also has a bit of a rocky delay-filled history. It's been in some form of planning (read: "we should do this sometime") since at least last March, but for whatever reason there wasn't really a call for coders until August, and the list wasn't finalized until I guess sometime in September. Initial setup and speccing happened on and off from there on out, and the first real code commit didn't happen until early November. It's been slightly rough going from there on, as we all had to get familiar with the combination of language/libraries/platform we'd ended up with, so small changes often turned into an entire evening of poring over a manual. We also went through at least four different image-handling libraries and discovered they all sucked, at which point we had to find something else and rewrite the existing code again. That's all more or less past now and things are speeding up a bit.
Since this doesn't seem to be actually mentioned anywhere, the current list of coders is:
thecrypto (illustrious leader)
indicoyote
latiass
lexyeevee
kalu
net-cat
verixDue to a variety of extenuating concerns, not all of these people are actively contributing at the moment. Who is and who isn't is strictly classified (and rude!), but I will say that I am, and I wish more people were. Some of us just have school/work consuming time like Gabe Newell on donuts.
(If I may soapbox for a moment: honestly, it rather sucks being expected to build a fairly large, vaguely backwards-compatible system that will make users of the old crumbling system happy with tools I have never used before.. and have that userbase constantly make Duke Nukem Forever jokes to my face. Thanks, furs, very inspiring.)
Anyway, current status of Ferrox right now:
- Registration and login and all that boring crap works, complete with an actual damn captcha, for all the good the stupid things do.
- Permissions system, well, exists. I guess there's not a lot to be said there.
- Admin panel is started. I'm basically making admin things up as I think of them, since I am not an admin and have no idea what current FA can and can't do.
- Gallery has basic functionality. Upload, look at stuff. Tagging/searching is roughly working.
- News and journals exist, can post/edit/view; no comment support yet.
- Clone of the current theme exists. Support for different themes and color schemes exists. There's also an alternate theme I made up, but I think I was trying to keep too much in the spirit of current FA and made it too bland.
- I'm doing notes right now; they're almost done.
- I'm trying to aggressively refactor and clean up as I go, so the code stays as compact and simple as possible. Always an uphill battle, but seems to be going okay.
3300 lines of code, 1100 lines of template, seventy or so commits.
Not much in the way of user-facing stuff (watches, profile, whatever) has been done, but a lot of that stuff is really just icing that can be banged out in a few hours. What I mostly want is a fast, flexible, and robust search backend we can use everywhere. Ah, I can dream.
Feel free to watch me; I will try to post something every week or so, and possibly throw up screenshots sometime. In theory, this will get my ass in gear so I actually have something to say every week. 8)
I might also brainstorm here if I don't know how to approach some feature; I'm not an artist, after all, so I am occasionally guessing wildly at how people might use this thing.
honestly though, gotta say thanks for all the hard work you and the coding team are putting into this:)
That's good to hear, though! It's too bad I can't help at all.
I'd like to explore into that more later, but for right now, I think it's a sort of low priority. But having the option for it would be a good idea.
Pretty much all that was ever needed was "taskforced", trustworthy individuals who could translate on-request a few lines of /depersonalised/ message at a high quality when there were specific issues or points needing to be made which required clarity of understanding. Worked AOK in the past for me for Spanish, French, German and Finnish without any difficulty whatsoever (helped to have at a degree of personal fluency, too, even if not, not in Finnish ^^).
Full "localisation" of FA is a bigger game but equally easy to deploy on a "taskforced" basis without any /specific/ need for further admins, as discussed in passing previously.
d. :)
Anyhow; you know we've got at least one totally trustworthy expert in the Russian department plus a few others who could translate on a /depersonalised/ message basis as-and-when who could be "taskforced" in on such a basis.
Only on very, very rare occasions is there a need for real-time communications with any community members in any language, including (American) English(!) and in most time-critical cases a stop-gap translation will suffice pending clarification.
Even if there is full "localisation" of text (to help ensure that community members are understanding the rules and thus saving time-and-effort for all), there is still no overarching "need" for a full admin team in each language as the default for general communications can still clearly be stated as English, with translation-on-request through to full capabilities depending on the general multilingual progression across the entire site/community. :)
Have you envisioned anything specific beyond that; or presumably still "back-burner"?
d.
p.s. The key word if any in that was "taskforce"; and, as noted various times before, that can be applied across a wide range of aspects within FA on a /non-admin/ basis.
p2.s. An alternative solution, given a 4 minute response time, might be for me to pay for you to go on a couple-dozen fast-learn language courses and have you personally available "plugged-in" to FA 24/7. ;) *kitty ducks incoming ^^*
Since the artists and writers will be the power users of the site I hope some of them are involved with alpha and beta testing. I've been in that spot myself: release a new fully functional piece of software to a customer base, and they immediately turn around and slam your hard work with, "Where's this feature?" and "Why doesn't it do this? We need it to!" and "That's not how we do it! Who told you we did it that way?" All because you failed to do proper research first.... ah, memories.
Cheers on getting some initiative on the Ferrox project, Eevee. Sounds like just sone basic lack of good management that delayed it from progressing further last year--no one took the reins, so-to-speak.
Ferrox is going to be grand...A search system, finally...
good job.
Y'know, I volunteered to help out with coding a long time ago... and was never contacted again.
How long ago? Last fall, or before that?