Helping

[liv]

Some time last week, I was in a bad mood such that what I felt like doing was spending hours playing simple, repetitive computer games. But honestly doing that does not really make me feel better, except very very short term, it just makes me annoyed at myself for wasting a lot of time on pointless things. So I had the brain-wave of deciding to sit down and work out what it is exactly that I crave from computer games when I'm in that kind of jangly bad mood. I concluded that I wanted to be doing something sufficiently difficult to give me a sense of accomplishment, but easy enough that if I concentrated I'd have a pretty good chance of succeeding. And I wanted lots and lots of immediate feedback. I figured out that a more productive activity that meets that need is programming.

So I took the plunge and asked to get my Dreamhack (DW development environment) reactivated after a three-year hiatus, and started working on a really small styles bug. And I jumped back into #dreamwidth IRC and lots of people were pleased to see me, and entirely willing to step me through fixing things when I messed up, and several completely vanished evenings later, I had a couple of patches accepted. The programming part was exactly what I needed, soothing and just challenging enough to be satisfying, and every time I changed something and compiled it and it worked, or tested something and found a bug and corrected it and tested again and it worked, I got exactly the hit that I get when I get a combo bonus on a silly computer game. Except that at the end I felt I'd contributed something useful by making DW better, instead of feeling drained and crappy because I'd wasted time improving my highscore on a game I don't care about.

The thing is, it's clear that the most useful thing I can do for DW is creating content here which attracts readers and helps to make it a lively community. I'm much much much better at that than I will ever be at programming! I also make a point of paying for my account, even though I don't need to, a free account would be perfectly adequate for the way I use DW. But I do increasingly like the freemium business model better than all the other alternatives that people are trying when it comes to monetizing the hosting of user-created content. (I am really not impressed at Flickr switching from freemium to ad-supported, ugh ugh ugh.) I sometimes dabble in documentation and I'm among the top 20 volunteers providing answers to Support questions. Though these days the site FAQs are good enough that there aren't many people asking questions simple enough for me to answer, people are posting to Support with actual bugs or very complex requests, and this is a very good thing, but it makes my contributions somewhat redundant.

So realistically, the little tiny bit of programming that I'm able to do is much more for my benefit than for Dreamwidth's. I am helping a bit, because there are always more tiny trivial bugs than there are people who have time and skills to fix them, so more hands on deck is not completely pointless. That certainly does contribute to the my enjoyment of working on the project; I have tried a few times to teach myself programming by going through exercises to write hello world programs and list-sorting programs and finding prime factor programs, and I tend to lose interest before I really learn anything, because I can't help finding the exercises pointless. Actually fixing something that makes a real-life project better is so very much more satisfying, even if it means I'm learning in a slightly odd order because I'm delving into a (very!) mature codebase and figuring out what to change. I like immersive learning though, I learned most of my Hebrew grammar through a really excellent course where we just started at the beginning of the Bible and went, ok, what's going on grammatically with this word, and got through about one and a half verses over the duration of the course. And working on DW feels very similar, it's real code, it's not example code, so yes, it's complex, but also it's valid in the technical educational sense.

Also, wow, I really do like the way Dreamwidth is completely reinventing Open Source culture. It's not a "meritocracy", it's a community where people take more pride in mentoring and training newbies than they do in competing to be the most l33t super-hackers. Some really really skilled people, including mark himself, held my virtual hand in IRC when I was getting frustrated with version control. And nobody even so much as hinted that there's a problem with me taking up several hours of senior devs' time in order to make a one-line fix.

This is the kind of reason why DW has a huge crowd of people working on it, some with quite a lot of responsibility for major projects, who don't at all come from the traditional Open Source / hobby programmer roots. Specifically, it has a clear majority of women, and greater than a random population sample representation of people with various other minority identities. I feel weird about the fact that I need a supportive of women-and-minorities kind of environment to be able to get involved in OS; I don't think of myself as the kind of person who needs that kind of "nurturing" to be able to learn. But the fact is that I gave up programming when I was 8 or 9, for reasons that do have something to do with being a girl, and I have a whole lot of catching up to do with people who spent their teenage years hacking for fun or pocket money, and probably their 20s working in some kind of software related field. And I don't think I could bring myself to do that in an environment where you get mocked for being a n0ob, or even where the atmosphere is friendly but competitive; obviously I can't compete with people who have 25 years more experience than I do.

This reflects the principle that I keep rediscovering when thinking about institutionalized bias: good practice helps everybody, bad practice hurts minorities more. I appreciate that some people actively prefer the OS world because you don't have to bother with being politically correct to the standards expected in the corporate environment, because your ability to code is the only thing that counts, even if you're not very good at social niceties. And that's great, that is indeed a really valuable thing, but it really doesn't lead to diversity; women are ridiculously under-represented in almost all major Open Source projects (apart from Dreamwidth and AO3), even compared to women's under-representation in technical fields in general.

A silly, personal example: I feel a bit self-conscious about the fact that Dreamwidth has now migrated to Git for version control, so that all my newbie mistakes are right out there on the public internet for everyone to see. But I have also read that this is a typical problem that women have with getting involved in Open Source. The real solution to this is to change the culture to give learners more privacy if they need it, but my immediate instinct when I read about something that women typically do that holds them back is to say, ok, I won't do that, then, I'll do this the way a man would. So I have determined not to worry about my stuff-ups being in my GitHub stream, because realistically nobody's even going to bother looking and even if they do, I don't care what some random stranger thinks of my ability to use version control. Obviously I'm bad at it, that's because I'm a beginner, not because I'm inherently incompetent.

The other reason this is important to me is to keep me empathizing with learners. mixosaurus made a really powerful blog post about respectful and compassionate teaching. Kat is right on the money that the sort of people who become academics may never have been crushingly bad at any academic subject, whereas by definition we're going to be teaching people with a range of abilities and levels of motivation, not just those who excel and love the subject and go on to become experts in our fields. It's really, really good for me to remember what it feels like to be a beginner, to be too scared of making mistakes to actually make progress. I know I have students who find, say, immunology as arcane and jargon-filled and arbitrary as I'm finding Git right now (like I said, the programming is going fine so far, it's the version control I'm struggling with). And they're not "stupid" and they're not deliberately refusing to try just to be awkward, they're beginners, or they're people who have succeeded at somewhat related skills (such as A-Level biology) but find that this particular intellectual field doesn't quite fit with the way their brain works.

So learning new skills is good for me as a teacher, as well as being satisfying. I've revived my dev journal at livredor to document some of my learning process, and since I have it I'll probably throw the baking in there as well, and maybe some of my very beginner-ish Arabic, though I generally know how to learn languages so I have less need to work on the meta-cognition stuff there. I strongly expect that 99% of people will find this detailed documentation totally boring, which is why I'm hiving it off into a separate journal. But I'm also mentioning it just in case you have the exact mindset where you find watching people learning new things interesting, and if you do you're welcome to watch the dev journal.

And if you have wishes for DW, well. I am not quite at the level where I can scratch my own, or my friends', itches yet, but I'm hoping to get there, at least if we're talking small itches. I mean, ideally what I would like to do is write a couple of smartphone clients and a decent front-end for the image hosting and a tool for exporting journals properly with comments and maybe something to import from Tumblr before Yahoo screw the site up beyond repair. But those are way beyond my capability and likely to remain so, because this is always going to be a hobby for me, I don't have a thousand hours to actually become an expert.

However, what I can do is a whole bunch of things related to the back-end that generates journal appearances. The most complicated things I've done so far are writing the Page Summary module, and writing code that mainly affects designers rather than end users to do with the option to colour-code entries on your reading page depending who posted them. Certainly, if there's a colour combination or a display option that you wish existed, there's a good chance I can make it happen. Note that I am not a web-designer by any stretch of the imagination, but what I can do is implement someone else's aesthetic concept in a format that can be made into an official style on Dreamwidth.

In order to be able to do that it has to be something with a licence that DW can use; if you designed something yourself you need to submit a CLA before it can be incorporated, if it's someone else's design then it needs to be public domain or under an appropriate licence. This means that I can't make official DW layouts with, you know, Game of Thrones wallpaper, and it also means that I can't make a Dreamwidth version of LJ Flexible Squares, which is a very common request. What I can do is take CSS-based layouts that modify Flexible Squares, and apply them to Dreamwidth's Tabula Rasa, because actually DW has more flexibility for styling with CSS than LJ ever did, it's just not very well publicized.

So tell me, next time I am in a bad mood and want to spend an evening programming, what can I do for you?

Comments

[rmc28]

Actually fixing something that makes a real-life project better is so very much more satisfying, even if it means I'm learning in a slightly odd order

This is what I love most about programming and wish I had had more access to when studying computer science formally. I'm so pleased to see someone else articulate it too.

It's also part of why I like working in "live ops" rather than "projects" in my current role: I really like making lots of small fixes and improvements that directly make people's lives (work or student) a little bit easier. Projects take much longer to get to the payoff and I prefer a daily fix (in both senses).

[liv]

Heh, conversely I'm glad that at least one real programmer feels as I do about what makes it satisfying. And it's cool that you get to do that at work, make little fixes and improvements with a fairly short-term payoff.

Also, I love your new grinning icon. (I haven't seen you in person since you cut your hair.)

[simont]

I have tried a few times to teach myself programming by going through exercises to write hello world programs and list-sorting programs and finding prime factor programs, and I tend to lose interest before I really learn anything, because I can't help finding the exercises pointless.

*nods* Me too! Whenever there's something new in programming I need to learn (usually a new language, or an update to the standard for an old one) I too find that the kind of noddy programs through which a deliberately designed course will walk you are just not motivating, and I don't really start to take things in properly until I have a 'real' problem to solve. Even if I make up my own exercise problems, they still tend to feel pointless in much the way you describe if they have no purpose at all other than 'demonstrate that I understand [foo]'.

I, of course, am one of those people who spent my teenage years hacking for fun – and one interesting thing about learning to program that way, in my spare time and at my own pace, is precisely that I was able to choose my own problems to solve rather than going through a pre-prepared set of artificial exercises. So even when the programs I wrote had no practical use, there was generally at least a frivolous use for them, e.g. "look at the pretty picture they produce" or "use them to startle my sister"; in other words, there was some kind of satisfaction to be derived from having written the program, which acted as motivation to write it.

It's a bit of a dilemma, really, because people who are trying to design teaching courses can't very well provide 'real problems' for everyone to solve – partly because any real problem ceases to be real (in the sense that motivates you and me) as soon as someone has solved it, whereas course materials would understandably prefer a reasonably static set of problems to the huge ongoing effort of constantly finding new ones, and also partly because by the very nature of real problems which haven't been solved yet it tends not to be completely clear how difficult they are or exactly which knowledge or skills they will require, so it's hard to arrange a collection of them into a sensibly paced course.

[liv]

Ooh, really interesting to hear about how you learn. (The more so since none of what I'm talking about would be anything like as doable without PuTTY.)

Definitely makes sense about real problems, even if they are frivolous problems. As for the issues that creates with designing courses, only imagine what happens when the "real problems" you'd ideally like to get your students to solve are medical problems, not just coding problems. Even apart from the ethical issues it's a major obstacle to giving students real-world experience when you don't know in advance how difficult it is likely to be to manage a particular condition or which knowledge and skills are relevant.

[simont]

PuTTY is kind of a special case, since a major motivation for writing large parts of that was "avoid revision" :-)

Now I think about it, there have been a few cases where I did attempt something moderately ambitious more or less just to convince myself that I could, with no real use for the result other than "feel satisfied at having managed it". The example that springs to mind is the first time I implemented a programming language, which – as is typical of one's first attempt – was far too trivial to be useful for anything. I suppose that kind of thing does count as a pure learning exercise of the kind I said I generally found unmotivating, but the motivation was that it was something I'd been specifically curious about for years.

Course design: yes, I can imagine there are other fields in which the tension between "good learning exercise" and "properly motivating problem" is even more of a pain! Though I also wonder: does it perhaps help the motivation question (not as much, but a little) if one is solving noddy-learning-exercise problems as part of an organised full-time teaching course, with marks and grades and not looking silly in front of your supervisor hanging on doing them well?

[liv]

Heh, it's very cool that PuTTY happened because you were avoiding revision, yay constructive procrastination!

I think a sufficiently hard problem might be interesting just for the sake of seeing whether I could solve a problem. I mean, jack and I intentionally spent half our weekend working on a probability problem someone posted on Twitter, which was totally abstract and pointless. But because I'm such a beginner, I'm not yet able to even formulate a problem which is hard enough to be a stretch for me without being impossible. Let alone actually do anything difficult. So at the moment I need the sort of easy yet meaningful tasks that Dreamwidth's effort-minor bugs provide.

You are absolutely right that instructor-provided feedback sometimes has to substitute for the reward of actually solving a real-world problem. It's something that teachers have to be really careful to do right, otherwise it's worse than useless, it just makes people scared of what mark they're going to get instead of being motivated to learn. Which in fact reminds me that I really ought to stop procrastinating and go back to working on my dissertation on just this subject.

[kerrypolka]

1. Dreamwidth is just great, isn't it? I've decided/discovered that coding is not something I enjoy enough to spend a lot of time on, but from all my friends at various development levels it just sounds like DW is a lovely environment.

2. In answer to your actual question at the end! I would love a way of making tags private, eg being able to tag my own posts to "THAT ONE COLLEAGUE I HATE" without having that tag appear anywhere to anyone who isn't me. I'm not sure if that's the kind of thing you're talking about - apologies if not!

[liv]

Yeah, if you basically hate coding, no amount of positive supportive environment is going to make that a fun thing to do with your free time. I really enjoy it, but I am frustrated by the learning curve thing, I can't yet program well enough to do stuff I would find really interesting, which in turn makes it hard to motivate myself to improve. DW is really helping with that, because even if what I'm doing is quite basic, knowing that it's helping the site is a big boost.

The tags thing is I think as much philosophical as technical. My understanding is that tags always have the lowest privacy of any post they appear on. So basically you can't have a private tag on a public post. There's kind of a high potential for drama otherwise. But yeah, I think a lot of people would like to be able to tag posts "that one colleague I hate" for their own use, and it feels like it would build enough on existing code that I could maybe do it. I will ask around, but my feeling is that it's not going to be encouraged as a feature.

[kerrypolka]

Is there a high potential for drama? I can't understand how (this is me not understanding, rather than scoffing, FYI!) - unless we're talking about levels of access rather than just private/public. I just thought would be nice to be able to tag posts eg "I should look up some porn based on this!" or "For the historical novel I'm writing maybe!" without everyone seeing that.

[liv]

There's always a bit of a problem where there's different levels of privacy within the same page; it's hard to rule out all the obscure combinations of circumstances that can lead to the private information leaking. Like, let's say someone makes a journal layout that displays entries in pink if they have more than three tags, purple if they have one or two, and blue if they have none. (I'm fairly certain this is possible, though I haven't tried it.) And if the person who coded the private tag / public post thing hadn't thought of that in advance, the multi-coloured style person could see that an entry has more tags than are visible to them. If you're really unlucky they could find a pattern in those entries and find a common theme between posts where you used a private tag.

I mean, maybe that's too minor an exception to be worth caring about, but DW is really really strict about making sure no information ever leaks about stuff that people have set as private or access-only. Which is one of the reasons that DW is not evil like FB is.

[synecdochic]

yes, what liv said -- there's way, way too many potential for information leaks that way, not to mention that we'd then have to come up with a UI for "create this tag privately" when you typed a new tag into the update page, plus a UI for "turn this tag private" on the manage tags page, and both of those workflows are already very complex, and more options are bad UI -- not that options shouldn't be allowed, but there needs to be a really freakin' good reason for them.

basically, there's no way to manage "private tags" that doesn't lead to a UI clusterfuck, which leads to people creating things publicly that they didn't want public. it is much, much, MUCH easier for tags to inherit the security of the posts they're used on.

workaround for tagging public posts with a specific tag that you don't want others to know about: either tag with a nonsense term that you know what it means, or use a certain word or phrase inside the post and search your journal for it.

[kerrypolka]

Thank you (both) for explaining that - I hadn't realised you could change post display based on tag number, nor had a clear idea of just how complex it would make the tagging experience for users. Thank you!

Private tags workaround

[(Anonymous)]

Option: you could use a completely different tagging system, like pinboard.in or delicious (although the latter has questionable stability), and tag your posts privately there (on pinboard.in/whatever) for those sort of posts, maybe. -- Selki

[green_knight]

have tried a few times to teach myself programming by going through exercises to write hello world programs and list-sorting programs and finding prime factor programs, and I tend to lose interest before I really learn anything, because I can't help finding the exercises pointless.

I am making the [retracted] effort to learn programming proper. (And really should make a proper post of my own.) But one of the things that help me this time around is that I'm doing everything until I have understood it, and in the process I am writing a thousand small applets for my own use, each doing a thing. Right now, I feel I need to learn at my own pace, so a project would be too much pressure. I think what we both have in common is the need to actually *program something* rather than merely typing out somebody else's code. (I've never learnt by just repeating the steps; I need to understand it - why would I choose this, what happens if I get it wrong, *how* do I get it wrong and how do I need to fix it, because _just_ doing it right - especially if half the time it doesn't work after all - isn't helping me much.


the option to colour-code entries on your reading page depending who posted them

What I would love is a better way to incorporate custom colours into the layout of a reading list. On LJ, I have a sidebar that shows custom colours for each person on my flist. On DW I have set custom colours, but right now they show up as a couple of pixels around the userpicture. And while I could work out which part of the CSS I need to modify to make that more visible, I would love the ability adjust this with clear instructions on 'add this tag in order to make this bit show up in custom colours'. If you're doing that, my eternal thanks will be yours. (and the ability to colourcode, say, a custom group would be icing on the cake.)

[liv]

That's really interesting about how you learn. I've enjoyed your periodic posts on the topic and I'll probably be even more interested now I'm actively learning myself. The thousand small applets each doing a thing sounds like a really good approach.

And really good point about programming versus typing in code. When I used to program back in the 80s, I did type a lot of code from magazines, but I always understood which bits I could change if I wanted to, silly things like giving myself 100 lives instead of 3 (I was a kid, after all), but it helped me to understand what I was typing. Nowadays, with online or book-based courses, they often only tell you whether you got the code exactly word-for-word right as the course author intended, so you have no idea if you just made a variant but still functioning program. CodeAcademy tries to be a bit fuzzy, but it is kind of buggy so often it was giving me errors for things I knew were right (just not identical to the suggested solution) or accepting absolutely any nonsense as valid code.

The custom colours thing: those options exist, but I didn't expose them to the end user. I made lots of options for layout designers exactly as you describe, but more experienced people agreed at the time that which bits have custom colours is a design decision, not necessarily a user option. Exposing at least some of that as an option where you can just tick a box in the wizard is technically doable, but I've made the system really quite flexible for designers, it wouldn't be very straightforward to translate that into nice simple tickyboxes. Basically you can make pretty much any element (or several) of the page have custom colours if you want to, so it's hard to condense that into wizard options.

What might work for you personally is a theme layer with the following code:
set custom_colors_template = "%%new%% .entry .footer { background-color: %%background%%; }
%%new%% .entry .footer a { color: %%foreground%%; }"

You can't do this with CSS alone because you need database access to be able to decide which person should display in which colours, it has to be in the code. Which is another argument for why it should be in the wizard, in some ways.

What exactly do you mean by "colourcode a custom group"? Do you mean, instead of having liv be one colour combination, and lethargic_man another, you can tie the colours to reading list filters, so everybody in the "SF fans" filter is one set of colours and everybody in the "random DW people" filter is another set? That would be a lot harder because it would require me to completely recode the way that custom colours work, but I can certainly see how it might be a useful feature when I get a bit more confident at programming complex things.

[green_knight]

The thousand small applets each doing a thing sounds like a really good approach.

I looked at the thing that all computer courses/books had in common that didn't work: the method of 'follow me along this path, and poof! you're a programmer.' And then I thought about the complex skills that I *had* learnt and came up with learning to play an instrument, where you start by doing small things over and over until they're second nature Think of learning to type 'adsdfa': you do it again and again until you can do it in your sleep. So now I'm creating variables and arrays and changing values and occasionally calling functions, over and over and over until everything I'm doing has become second nature and I no longer have to think about it. And I'm very much at the start - I am currently working on truly grokking opening and saving files (there are some surprises left) - but I've solved my first couple of challenges, and I feel that it's sticking. Go me.

And thanks for the code snippet - I shall investigate more when I have the time. The custom group colour probably is something I should post to suggestions if I ever get around to *that* - because you'd need to tag is somewhere or otherwise devise an interface for 'use this colour for this group.' But yeah. When you're not made to feel stupid for not knowing everything yet, programming is a lot of fun. Who'd have thunk ;-)

[vatine]

I have recently been writing C++ for work, mostly motivated by "it needed done" and "no one else has time" and so 20 years of not learning the language has been crushed, like a can in a recycling centre.

I still don't know what a good way to learn "programming" or "a new programming language" it's, but a c combination of "achievable" and "requires some stretch" is probably a good starting point.

[liv]

Heh, I like the image! For me I find it relatively easy to learn programming languages, you can always look up the syntax after all. But learning to decide what exactly you want your program to do, and how to structure it, that's where I'm on a very steep learning curve just now.

[vatine]

There's some element of that. I usually find syntax pretty easy, but semantics and idioms are also required (otherwise my code reviewers complain until I fix my code). :)

[green_knight]

learning to decide what exactly you want your program to do, and how to structure it, that's where I'm on a very steep learning curve just now.

Well, yes, that's the _programming_ part, as opposed to 'learning the syntax.' And there's very little advice on it, and at least in Mac Programming half the people who are writing about it are thinking in different idioms than the one they're actually using, which does not improve matters. Every now and again you catch them out when you see the hoops they jump through to make the system perform to their expectations, but you need to already understand a lot of stuff before you can spot it.

For me, the most useful book was one on programming patterns - high-level abstracts that told me something about the toolbox of my chosen language (Objective C). After a year of intense reading and engagement I am just starting to see simple solutions to problems and to develop a feel for implementations. Very often I want to do things as I would have implemented them in Filemaker, which is a completely different idiom, but I am starting to think Objective C. It's cool when that happens!

[forestofglory]

Hmm ... I really want to see how your baking is going, but am not much interested in the programing stuff. I suppose I shall follow Livrdor and skim the stuff I don't care about.

[ptc24]

The thing that has changed since the old 8-bit days (incidentally, I once saw a graph of gender ratios of computer science undergrads over time; before the home computer generation, the ratios were a lot better than they are now) is the difficulty of doing things. With an old 8-bit home computer, making some graphics appear on the screen, and making it respond to the keyboard, was pretty easy; there wasn't really a distinction between the graphics commands and the "core" of the language.

These days... your core language might be nice and easy to learn, but if the libraries and APIs you need are difficult, then the learning curve becomes awkwardly shaped.

Anyway, I had a thought about archiving, as it's odd that people don't seem to bother with the comments. I had a look - the API documentation is really quite rotten - and found something. Try this URL: http://www.dreamwidth.org/export_comments.bml?get=comment_body&startid=0

Now if you view the page source, and then save the page source, then you've got a nice file. I don't have a lot of experience with teaching programming, but it looks like the sort of file one could set a lot of very nice pull-something-out-of-this-file exercises on. Put it like this: if I see a file like that at work, I smile and say, "this shouldn't be too hard".

[ptc24]

Oddity of the day - if you click the link, it does the wrong thing. If you paste it into your browser, then it works nicely.

[randomling]

Wow, that is weird.

Wandering off to investigate...

[synecdochic]

IIRC, it does referer checking so that people can't call it outside the DW context to get other people's comment data. i may not be recalling correctly, mind you! but i vaguely recall that being the reason.

[synecdochic]

the API documentation is, wholly and without a doubt, utterly awful. (the APIs themselves are probably in mildly crap shape, too; we've tried to keep them up to date as we've changed things, but undoubtedly they've bitrotted a bit, especially since we've had our hands tied some with needing to stay reverse-compatable with LJ.)

having someone adopt the APIs and run with them is the sort of project i'm abso-fucking-lutely and eternally hopeful someone will pick up someday, especially since it's the precursor to so many things, but it's one of our areas of massive technical debt and there are very few people who are both interested in picking it up and possessed of the experience necessary, and the people i'm allowed to order around because i pay them are overbooked like whoa and working on other things :/ it's the problem with running with a nearly-entirely-volunteer workforce, especially one that's so newcomer-heavy. (i mean, don't get me wrong! i think it is awesome how many newcomers we have, and how many people find us a supportive and encouraging environment to learn in, and i am perpetually proud of the environment we've managed to build. but it does make it harder for sweeping projects to get implemented.)

which is all to say, if you (or anyone else reading) wanted to adopt the cause of hauling our APIs into the 21st century kicking and screaming, the response would be a very enthusiastic "yes, please".

[green_knight]

having someone adopt the APIs and run with them

Yeah, I've looked at them, but I understand far too little about APIs and I could not get through the documentation, so I had to step back again. I'd be happy to contribute, but on my own, I get lost extremely quickly. I don't even know what a good API looks like...

[ephemera]

ohh - interesting. I have a very similar-sounding self-soothing-with-puzzle-games pattern going on, so I am now wondering if maybe it would be worth me re-visiting (*again*) the whole learning to program issue. (I keep trying and failing hard at the step where you move from assembling things out of pre-made blocks to the nebulous protean soup of raw code.)

[ceb]

Go you! I am sriously impressed that you've managed to channel your time-wasting into such useful things.

[momijizukamori]

Heh, I actually found I got the same sort of emotional 'hit' - I did something, it was useful to other people, and my brain mentally filed it as 'productive' instead of sitting around reading reddit or whatever. And while you may think the bugs you're fixing are 'trivial' and don't really matter, I will say right here and now that it matters a lot to me! In part because I am on an endless quest to beautify the site more, and in part because I'm one of the (now three with you) people working on styles patches, it's stuff I don't have to do. A lot of our new devs hop in and start poking around at the perl back-end stuff while I sit and go 'nooooooooo come to styles we love you /;o;/'.

I also have similar problems with trying to learn programming - the artificial progression doesn't engage me, particularly as I've now tried enough times that the beginning steps ('this is a string, this is an integer') are meaningless to me. When I first taught myself HTML and CSS (which is what I spent my preteen years doing, instead of C), I did it because I had a goal in mind, and at the time was young and eager enough to be impressed that something worked, rather than automatically comparing it to 'better' sites. I'm trying to work with the 'have a goal' philosophy for new stuff - it worked for S2, which I've mostly learned via a combo of Dre asking me to redo site-scheme icon pages so they were table-less, and then deciding to tackle 'kill talkread.bml in fire'. Perl is proving a little trickier, I think in part because DW's perl codebase is HUGE so there's a lot of 'what is this variable, what is it doing, help I'm confused'. But I am determined to get a better styles sorting thing working sometime soonish.

[jewelfox]

If you just want a personal account, you can get a private Github if you are female. >_>b Or something like that.

I agree very much about the "meritocracy" crap, and it's part of the reason I left the Linux world even after interning at GNOME. In some ways especially after. I got trolls on my journal any time I talked about it, or feminism, and after I spent awhile both advocating a particular change on Planet GNOME (making JavaScript the official "app developer" language) and writing developer documentation towards that end, they came out and made exactly that change and no one seemed to remember I'd said anything about it.

Er, sorry. >_>;;

This was a very good essay, and I agree with a lot of it. I'm trying in between depressive episodes to write a Windows Store app for Dreamwidth and would be happy to compare notes or possibly ask for advice if you're giving it out. >_>b

[zvi]

dunno, if this was the sort of thing you were thinking about, but if you wanted to reverse engineer zesty to working as an official style, that would be a good thing.