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

Friday, June 27, 2014

Why is there no "Ravelry for Sewing"?

Ravelry is possibly the most loved web site on knitting and crocheting. It is a site that, since its inception in 2007, quickly grew to become the first stop on the web for information on knit and crochet patterns as well as yarns. However, it is much more than that: It is a social community that allows users to form their own groups. It is a place where people can keep track of their knit and crochet projects - as well as share experiences and pictures related to these projects. Built on its huge, user-contributed pattern/yarn database, the site offers a tool, called 'stash', that helps knitters and crocheters organize their yarns and patterns. Beyond that, Ravelry is a place where pattern designers can sell their patterns as PDF-files right through the site.

Unfortunately I personally only learned about them very recently - I missed the opportunity to watch them grow. Still, despite not being a knitter or crocheter at all, to me the site feels like a beacon of sanity in an insane world.

In the rest of this post, I share my observations on
  1. What I think are the main reasons that contribute to Ravelry's sucess, and
  2. I discuss some possible reasons why there is no "Ravelry for sewing" yet.
I already know that, in the first iterations of this post, I will likely not be able to provide a comprehensive list. However, if you have something to add, please comment. I'd be glad to incorporate your comments to compile such a list.

Things that Make Ravelry Great

  1. It is built on mutual respect, a small set of sane rules is enforced.
  2. It has a well-structured database with links between database entries,
  3. It has a powerful search interface, 
  4. It gives users the ability to edit the database (with very few restrictions),
  5. It generates so much value from its database that its members actually add and improve data,
  6. It gives designers useful tools for promoting and conducting their business in the appropriate setting,
  7. The ads displayed on it are tasteful and interesting,
  8. Its forums are simple to use and unobtrusive,
  9. It favors usability over design,
  10. It's a down-to-earth family-operated business.
(Sorry, some of these are pretty subjective.)

On the one side, Ravelry has rules against unsolicited advertising in the main forums and in database entries. On the other side, Ravelry requires that critique should be given to the owner of the discussed item first - and that it should be constructive. There are strict rules against libelous comments - posters must back up their claims with facts. (While there are laws against slandering others in many countries, when one looks at what's going on on many web sites, one can easily forget that these laws exist. It's refreshing to see that the founders of Ravelry have achieved the not-so-small feat of building a community that, at large, cares.)

Looking at Ravelry's yarn and pattern database, one sees that it does not only contain the yarns and patterns themselves. It also records publishers (of patterns), sources (of patterns, e.g. books, magazines, ...), yarn brands, pattern designers, local yarn stores, and several more. At first this might look overly complicated, but in reality it's not - it's simply modeling the situation we have in the world out there. Organizing and recording all this data is the prerequisite for linking things: Pattern sources contain one or more patterns. Patterns can be contained in sources. Yarn often has a brand, and people can link the local yarn store they bought their yarn from in their yarn's stash entry. Projects can be linked to patterns. What I'm preparing to get at here is that Ravelry is organized pretty much like a spider web. Nearly everything that can be linked is linked and you can traverse this web of information easily by following the links between items.

Since the database items are structured so well, you can easily search for yarns by their fiber content, color, brand, gauge, recommended hook/needle size, and many many more aspects. Despite offering so many ways to filter the search results, the user interface is pretty intuitive and has a clear and consistent layout.

When you find a record that is not quite perfect, Ravelry allows you to edit and improve it. In that sense, it's similar to Wikipedia - anyone can edit, and if someone makes a mistake not all is lost because the site remembers all earlier versions of a page. Changes can, if needed, be reverted. However, some database entries, namely those where there is a dedicated owner, cannot be edited freely.

Ravelry actually provides people a great incentive to add data to the site: When you want to record all the patterns you own to keep a complete inventory of what you have, it's quite natural to just add the patterns the site doesn't know yet to the database. The site lets you link any pattern in the database from your personal stash. By offering the stash tool for free, Ravelry encourages the users to contribute to the site's database - and the users actually gain something useful in return.

Independent designers can open their own sub-communities (called 'groups') to keep in touch with their customer base. For free. They get their own discussion board, wiki pages, etc. They can sell PDF patterns directly to Ravelry members through the Ravelry pattern store. They can pay for advertising on the site at affordable rates.

Ravelry's ads don't blink, don't automatically redirect, and, resultingly, don't cause epileptic seizures. The ads are not littered all over the place, they have their dedicated, unobtrusive spots. Best of all, the adds come mostly from small business owners in or adjacent to the knitting and crocheting space.

Similarly, forum posts don't blink, there is no armada of emoticons hopping all over the place, the forums remember up to which point you've read, and they're easy to use. While this may sound like a really weird thing to say, I think this strict banning of blinking stuff seems to attract a very friendly and, possibly, more mature and experienced crowd.

One pattern that becomes obvious when using the site for a while is that the focus lies foremost on usability - design comes secondary. Links are highlighted on mouseover, boldly enough to clearly communicate that this is something that could be clicked. Important buttons are placed prominently. Default settings, like the number of items displayed in search, or the style in which the search results are displayed make a lot of sense. Different aspects of a page, such as the comments or projects attached to patterns, are organized neatly in tabs. I'm pretty sure there are more aspects that I don't even realize consciously.

My last point on the list is that I personally like how Ravelry is a small, family-operated business - and that they intend to keep it that way. Looking at the size of Ravelry's user base and the amount of page views they have, it's pretty obvious they could cash in big on what they built in several ways: They probably get offers to sell the whole site, they could run (the annoying) ads from big ad networks that pay more, they could start charging for their free services... But they have decided to go their own way that feels right to them - as a small team with a generous and friendly mindset. And I think that, simply by doing that and by being who they are, they attract a similarly nice crowd.

Why There is no "Sewing Ravelry"

  1. "Ravelry is gigantic",
  2. "There are too many sewing communities already",
Ravelry as it is today is gigantic, both in terms of its user base and in terms of its technical complexity. While it has been built mostly by a single developer, it's obvious that he's both competent and that he spends time consistently improving and extending the site. Anyone who steps up to the challenge of making a Ravelry-like site for another craft has a hard time comparing favorably in several regards:
  • Starting a new site with a single dedicated full-time developer means it will take a long time to get the site to the point that it doesn't look completely bad compared to Ravelry. Starting a new site with several dedicated full-time developers means that there's more mouths to feed and, thus, the site needs to make more money - which usually means more aggressive ads, higher commissions on pattern sales, and all that comes with making the step from a "small family-operated business" to a "niche startup".
  • Ravelry's pattern+yarn database is huge. Building a similar database for sewing patterns and designer fabrics needs both time and people who are motivated enough to do it. Since there already exist several databases for sewing patterns, it's likely not totally easy to motivate people to enter information to a new database. While the existing databases all have shortcomings regarding their searchability, these shortcomings could be remedied by their operators by changing the database structure - still, adding the missing records and links will nonetheless require a huge community effort.
People who say that repeating what Ravelry did is impossible often say that "there are too many sewing communities already". Maybe they are right, I do not know. I don't know how the situation with knit and crochet communities was when Ravelry appeared. Were there really no significant knit and crochet communities?
It's definitely true that there is a bunch of sewing communities that are pretty large (but still small compared to Ravelry). And people get used to their communities - both in terms of getting used to the web site's layout as well as getting used to the people they encounter there. People who are already busy within their existing communities might just not be interested in joining another one that is, in the first years, not as polished as the existing ones.

Actually, the ultimate reason why there is nothing that compares to Ravelry for the sewing world is that no one made it. Several people tried, but the undertaking is so massive that it's very very hard not to get discouraged at some point on the way. The basic recipe seems to be:
  • to have a full-time developer available to work on the site
  • to keep managing the budding community
  • to understand how people think and feel about things, how they organize their stashes, how they search for patterns, etc. - to a large degree, inspiration can be taken from Ravelry, but there is a good amount of things that need to be pondered, analyzed and adapted to properly capture and link sewing-related information
Following this basic recipe is surprisingly hard. For a single developer, building something similar to Ravelry is a monumental task and requires someone who doesn't resist learning all of the different technologies that are needed to build such a site. Keeping a sufficiently capable developer available to work full-time on the site means that it's important to make sure the site quickly earns enough money to keep them from having to take on a regular day-job to feed their family.

Good community managers are hard to come by. It takes someone with a good judgement, a way with words and people, the ability to deal with drama and resolve conflicts, and an overall helpful and positive attitude. Not everyone can do it (at least I'm pretty sure I can't). As the community grows, the community manager also needs to recruits volunteer moderators from the user base.

Understanding how people think and feel can actually be quite difficult. Yet it's very important for everyone who creates a web site. It doesn't help if you have a site that can do anything, has the greatest features and all that, but somehow has a design or layout that irritates and confuses people. Likewise, you can have site that feels good to use, has great people on it, but the technical foundation just doesn't allow you to do what you need to do. Both these sites don't realize their potential.

I think that the reason why we don't have a "Ravelry for sewing" is that a lot of qualifications and dedication needs to come together in a very small team with a very specific set of moral standards. There definitely are people who tried, or are still trying.

Actually, there is a wonderful blog post on just this topic already. I just wanted to add my personal view.

What's your view on why there is no "Ravelry for sewing"?