If you'd like to use the site and help me test it, you can join here.

Friday, August 29, 2014

Community-Edited Databases - Libraries of the Internet?

So, first off, I have to say that - as a kid - I used to be the person who could be called a bookworm. I was reading all kinds of books all day long if you'd let me. Apparently, I even spent a large part of one of my birthday parties sitting under the table reading a book while the guests were out playing in the garden. Yes, I'm pretty weird. :)

So, I really like libraries - I spent a good deal of time reading their books. Then, the Internet came, and I was a child in a family of early-adopters - so I actually had access to the Internet. I was amazed and captivated.

Today I'm still amazed - and today I want to tell you about my take on community-edited databases. In some regard, they remind me of libraries: They categorize, they catalogue, they provide links, pointers and information. Similar to the monastery-dwelling scholars of old times, we now have people on the web who - on their own free accord - take up these important and useful tasks of bringing order into a mess of data. And I think that's amazing.

In http://mmitscotland.wordpress.com/2012/02/07/5-reasons-why-we-really-need-librarians-and-information-professionals-in-the-internet-age/, Louise Morrison argues why libaries/librarians are still needed, even with the Internet. I can't but agree with her.

Why is it that Wikipedia or Ravelry are such huge successes? It's because they fostered a culture of responsible editorship and it's because they made clear that they will always be free to use.

Why is it that many database sites fail? It's because
  1. they failed engaging the people who feel joy and purpose in categorizing and ordering data
  2. they failed in setting up the site in such a way that their volunteers can actually contribute easily
Am I making any sense?

Wednesday, August 27, 2014

I want a "Ravelry for sewing" - and the only way to get one seems to be to build it myself: Kaava - A New Sewing Stash Database Site

Aug 27th, 2014: It's been a while since my last post and that's due to me (+ family) moving from a small apartment into a bigger one. Since a few minutes, we have regular internet access again.

Nov 27th, 2014: I'm still here, things are moving forward and are starting to become useful. There's still a good deal of rough edges and unfinished parts, but things are definitely getting better - step by step. I'm fixing and improving what is brought up by people as important. I'm thankful I'm not alone in this. Thanks to the feedback it's easy to stay on course and do something useful every day. Basically, the plan at the moment is to add convenience features, improve the existing features and to tackle site design (which I'm a bit scared of because this is the first time I'm doing web design).

June 30th, 2015: In the beginning of 2015, I wrote terms of use and a privacy policy (not from scratch, but using some templates). Afterwards, the site went through its first redesign. Reason being that I was using too many third party libraries. I started from scratch, writing my own CSS/HTML instead of using heavyweight frameworks that were too big for me to handle. There's still a few third-party libraries I want to get rid off, but things are looking a lot easier and more maintainable. Now, in a lot of places where the site would lead you to a separate page with a form to edit your data, the editing can be done inline, right on the page. The design isn't particularly pretty, but at least it's consistent and easy to change. So much to do, but we'll get there, step by step.


Today I want to tell you that I think I'm crazy enough to join the crowd of people (yes, in the sewing world there already are a few sites) who build databases for patterns/fabrics/etc. Personally, I can't imagine living a "normal life" with a "regular job" and I'd so much love to see a sewing database that really, actually, works. And if, in a few years, I can use that site to learn how to make clothes that actually fit me that would be awesome.

Obviously, I can't be sure I'm the person that can make such a site work before I honestly try. Since April, I've been hacking away learning web development and figuring out a decent way to set up the database structure. In June I started talking to a bunch of really nice people from the "Sew Obsessed" group at Ravelry. They've provided so much useful input and dispelled a bunch of my false beliefs that I realized that there's no way I can build such a database while keeping all to myself until it's "done".

The reason I, as a non-knitter/-crocheter, found Ravelry was this blog post: http://sewaholic.net/a-cool-new-sewing-site-kollabora-plus-a-contest/. Well, actually it wasn't the blog post itself but the wonderful comments that people made. I'm really grateful I found this post when I searched for sewing sites.

So, I'm building a site. It's not done yet - actually, it's in a pretty unfinished state. I currently have rough sketches of
  1. a pattern database (with pattern sources and publishers)
  2. a designer fabric database (with brands)
  3. a thread database (with colorways)
  4. a local sewing store database
  5. projects that can be linked to patterns
  6. a basic stash management tool
  7. a basic measurements application
Sep 27th, 2014: I've found that the best way to get this to a point where it becomes useful quickly is to start out with the pattern database (patterns and pattern sources) + projects + stash (patterns, sources and fabrics) as core features. So, the fabric database and thread database are gone for the time being - to make a reappearance when the rest of the site works well enough. Since my focus is on dressmaking, there's no reason yet to compete with the existing fabric databases. I think it would be awesome to have a site that allows you to see what kinds of fabrics others used in their projects for a particular pattern - and that lets you browse for projects that use similar fabrics to those you have. So, fabric-project-linking turns out to be a top priority for the site to launch with.
The site is barely functioning and design is mostly missing. Many things need structuring, so it's really a construction site. There's a bunch of things but all of them are pretty unfinished. I think that understanding the proper structure for a sewing database is something that should be done early on, before people add a lot of data and we all suddenly discover that things don't work out well and deep changes need to be made in retrospect.

I'm gathering all my courage to show it to some people so they can comment and tell me about things that don't work or need to be changed to work better. If you feel you could be such a person, drop me a short eMail. I'll let you know how and where you can access the site with a test account.

The reason why it takes me so much courage to do this is that I'm not really an experienced web developer. I spent the last years of my life doing research on formal verification, hardware construction and other obscure computer science stuff. All I know about actually doing web development I learned since I started with this in April. I'm learning new things every week I keep working on the site. Yes, having a general computer science background does help learning all these things pretty quickly, but it's no substitute for actual experience. Basically, this means that, if you can precisely say how something should behave, I'll learn how to build it that way. Learning something new is always easier when you know the precise result you want to achieve.

Then, there's web hosting which I've never personally dealt with before either. That's the reason why there's just a test account right now. I can't let anyone make accounts with their own passwords before I haven't consulted with someone who knows more than I do about setting up secure web hosting and until I have learned enough about that topic.

(Edit Aug, 28th, 2014: Actually, the site is behind SSL with a self-signed certificate at https://kaava.net now and I can let you have an account with your own password. Still, make sure you don't use a password you use anywhere else. This goes for every web site in the world out there, btw. - I wrote most of this post in July before the move and some things have changed since then.)

Still, I have made something that I think might show a glimpse of how things could be - even if it's not yet in a really usable or final shape. If you join me in this effort, it means that I will listen to your ideas on how things should be. You can help shape the future of the site by warning me of things done badly (by me or by other sites) and by pointing out things worth copying (conceptually, not literally) from other sites. Since people do sometimes have contradicting opinions and interests, I can't promise I'll do everything you say - I'll listen to everyone and take your views into account when deciding how to develop the site in such a way that it works for most people who are into dressmaking.

When you look at my other posts, you'll find it's obvious that I'm seriously impressed by Ravelry. They chose a path where they truly put the users of the site first (and it shows in their advertising model). Obviously, for a new site any such path will be a different one because the situation is different - but the only way to find and follow that path is through the people who feel they could use the site.

Drop me a short note to take part in this journey. I'd be happy to have you on board to make sure we get a useful sewing site up.