Code As Craft

Comments (Archived):

1. Another one is Facebook Engineering’s Notes: http://www.facebook.com/not…Something that shouldn’t be overlooked is the contribution a lot of these companies are also making back into the software development community by either contributing to current open source projects to make them better or opening sourcing their internal tools. Twitter and Facebook both have been very good at this:http://twitter.com/about/op…http://developers.facebook….

   1. Thanks for the link to FB’s engineering blog. I should have included that in the postTotally agree about all the open sourcing everyone is doing these daysIts awesome

2. For any coder types, you may find value exploring the hacking tag on my blog. I’ll see if I can have it go to a preview mode for tags so you can just see the top of each post.Excellent, the evermore wordpress plugin does precisely what I wanted. The previews are live.

3. I think there are a number of engineers writing about software development as a craft, rather than just the intricate details of particular software issues. Some of my favorites:Jeff Atwood: http://codinghorror.comJoel Spolsky: http://joelonsoftware.com (though he covers more entrepreneurial topics than pure software)Scott Hanselman: http://www.hanselman.com/blog/Landon Dyer: http://www.dadhacker.com/blog/Joe Armstrong: http://armstrongonsoftware….Wil Shipley: http://wilshipley.com/blog/Earl Beede: http://forums.construx.com/…Steve Yegge: http://steve-yegge.blogspot… (now inactive)Yossi Kreinin: http://www.yosefk.com/blog/

   1. Thanks for the list. Wild that I only had discovered Joel and Jeff. You reminded me I follow some folks on twitter, but people tend to talk about many varied topics there so post list filtering is good (and something I’m working on).

   2. that’s a great list dennyand what is exciting is that companies are now allowing their engineers toblog about their work on the company blog

   3. I would say that you can track the recent code as craft movement even farther back than this list of bloggers. It seems like the spark started at least as far back as 1999 with the Pragmatic Programming book http://www.pragprog.com/the… It should be required reading for software developers.

      1. Thats a great book! I went back to re-read it last year.

      2. Adding to the evergrowing ebook pile.

      3. I’m not a coder, at least not since the mid 80s. But I think I’ll read that book. Thanks

         1. The “Pragmatic Programmer” book was released under a new title: “My love-hate relationship with programming: how pure beauty can turn into a frustrating hell, and visa versa”

      4. At Etsy, I give a copy of the Pragmatic Programmer to new engineers on their first day. Quite often, they already have a copy, so I ask them to pass it on. It can only make the engineering world a better place.In addition to all the people who write about software as craft, we are also inspired by the “hacker ethic” (http://en.wikipedia.org/wik… ) that was first fully codified by Steven Levy in his book Hackers, particularly “you can create art and beauty on a computer.” Levy’s book is where I learned that the term “hacker” originated at the Tech Model Railroad Club at MIT in the 1950s: http://tmrc.mit.edu/hackers….

      5. Another great book for programmers is Code Complete http://www.cc2e.com/. The Pragmatic Programmer is often considered a compliment to this book, though both overlap on many topics. Code Complete is a massive book, roughly 3x the size of Pragmatic Programmer. Your developers no doubt know of this book already (especially if they are senior- or architect-level developers), and many consider it a must-read. I certainly do.Fantastic list, Denton. Other great blogs from developers I would recommend are:Simon Willison: http://simonwillison.net/Scott Schiller: http://www.scottschiller.com/Kent Brewster: http://kentbrewster.com/Isaac Schlueter: http://foohack.com/Nicholas Zakas: http://www.nczonline.net/Peter-Paul Koch: http://quirksmode.org/blog/Dion Almaer: http://almaer.com/blog/

   4. Peter McBreen’s “Software Craftsmanship – The New Imperative” book was important in making this meme explicit.http://www.mcbreen.ab.ca/So…Mark Bernstein’s “NeoVictorian Computing” series is a more recent example.http://www.markbernstein.or…

4. Cliff Click Jr.’s Blog: http://blogs.azulsystems.co…He was the “lead designer behind Hotspot’s server JIT compiler” and is now the Chief JVM Architect at Azul Systems, where they run Hotspot on top of their custom seriously parallel SMP processor systems (hundreds, up to nearly a thousand in the current 3rd generation).

   1. Thanks!

5. Good post Fred. I think that there is a wide consensus among agile software engineering community that we are a craft. The notion of evolving the right system through sculpting and adaptation has been triumphing over the dated water fall, so-called upfront engineering approach.Modern software languages and abundance of abundance of high quality free libraries make it possible to dramatically reduce software teams and deliver better software, faster. It really did become a craft of a few smart people as opposed to massive size projects with dozens mediocres.Here is my detailed post on the topic:http://www.readwriteweb.com…

   1. What languages best jive with that way of thinking about coding?Python, Ruby, Scala, PHP, Javascript?I keep thinking about coding abstracted out to physical interaction and gestures but I haven’t quite got my head around how we’d setup the detailed connections/functions.

      1. I’d have to say Java, but I am biased. C++ was really hard to refactor, because of the language complexity, but this day and age it boils down to tools. Specifically, many modern IDE ( coding environments ) have built in refactoring and unit testing which makes it easy to adapt and sculpt the code, while greatly reducing risk of errors.

         1. My background is C++, just last year I dipped into more web friendly languages. Now as I think about interacting systems the web itself appears as a the ultimate language, of billions of interacting interfaces with dozens of languages and hundreds or thousands of protocols.

   2. I’ll read that post when I get home alexDoes AB ever post about the coding work you and your team do?

      1. Yes we have done this extensively as well, more so in the past than recently. Here is a widely sited post from our blog:How and why we are using Amazon SimpleDB instead of Relational Database:http://blog.adaptiveblue.co…

         1. thanks. i wish i had seen that post when i wrote my post. i would have included it

   3. Like Alex, I see a lot of this in the Agile world. Bob Martin’s keynote at Agile 2008 was about the vital need for craftsmanship, for example, and the first Software Craftsmanship conference was held right next door to the big Agile conference in 2009.Readers here should check out the Lean Startup movement, a post-Agile process that combines Agile development methods with an exciting, startup-specific approach to business. One of the major things that has pushed against software craftsmanship was industrial-age approaches to business. Both Agile and Lean philosophies reward a craft orientation, which pays off especially well for startups.

6. We’ve been publishing a mix of articles, interviews, and API tips at the Flickr Code blog for a couple of years now.http://code.flickr.com/blog/A couple of recent good articles:A piece of our no-branches-no-tests-deploy-from-trunk development methodologyhttp://code.flickr.com/blog…What we’ve learned about i18nhttp://code.flickr.com/blog…A look at what *can* be done with MySQL (in the NoSQL vien)http://code.flickr.com/blog…The importance of funhttp://code.flickr.com/blog…Also at the bottom of http://code.flickr.com we publish when our last deploy was and who was responsible for the changes that went out so you know who to yell at if Flickr ever breaks 🙂

   1. Oh sweet, I did not know about Flickr Code blog, thanks!

   2. i should have known about the flickr code blog, but didn’t. thanks for pointing it out Kellan

7. I’ve got to give a shout-out to my engineer Dan, who has done an amazing job building http://teamhomefield.com completely solo. You can read his blog at http://danspinosa.comAs Fred told us, our company is “a little light on engineers.” He’s right, but Dan thinks of it as a positive. http://danspinosa.com/post/…That being said, we will be hiring a couple great engineers later this year.

   1. +1 to your shout to Dan…I’ve done a little code chatting with him and I also tend to prefer to get things off the ground and to a certain point in the code myself so I’m probably a little biased…but still, he’s def. a resource more people should know about and pay attention to (and I think they will as you guys continue to make waves) 🙂

8. Yes, the ‘Fail Whale’ story is pertinent: We’re well into building our web site, and in principle such things could happen to us.Can get a one step better approach borrowing from, say, ‘assembly line balancing’. A little more would be to regard the Web site as ‘network queuing’, that is, as a ‘network of communicating servers, each with a queue of input work to do and output work to transmit’.For such progress, first-cut, need to identify the ‘servers’ and communications paths that determine the network and get real time data on the queue lengths and service times.If as a whole the network has too little capacity, say, in the sense of too many ‘fail whale’ screens sent to users, then looking at queue lengths, service times, and time-outs should identify the ‘bottlenecks’. For a crude first-cut, an approach is just to alleviate the bottlenecks.To do any such things, would want good ability to collect data on queue lengths, service times, CPU busy fraction, page faults per second, page file size, etc. So, want some ‘instrumentation’. Nicely, there are now good tools for instrumenting the code and collecting the data.We haven’t built in such instrumentation yet, do believe we can ‘add it on’ routinely, and are toying with when to ‘instrument’ the code, before or after ‘going live’!For more, there is a network queuing optimization problem. There is a classic case: Say need 10 servers in the network, that is, 10 ‘nodes’ in the network. Suppose have 30 old computers, each with different processors, disk drives, network interface cards, etc. For each of the 30 computers, see how fast it can do the work of each of the 10 nodes. Yes, would help to have enough ‘scaffolding’ and ‘instrumentation’ in place to make this data collection easy! Then how to assign 10 of the 30 computers to the 10 nodes to maximize the capacity of the whole site? Some people attack this problem with simulation — simulation of queuing networks, including for designing computer systems, is an old subject. I have not thought about the fully general case, but for one important case get a ‘bottleneck assignment problem’ which has a nice solution (the solution actually is ‘polynomial’ and not NP-complete!).Then, suppose have the site all up and running. It’s going along just fine. Then start to get symptoms of problems — getting ‘fail whales’ too frequently on the screens of own ‘test robot users’, get e-mail messages complaining, maybe even get phone calls.So, how could we detect this problem earlier, automatically? Then how could we make progress on automating the diagnosis of the problem?The Twitter people did notice that they had a lot of data on a lot of variables. Yup!For the detection, say, 1000 times a second, ask if the site is sick or healthy. There are two ways to be wrong: (1) Conclude sick when healthy (false alarm). (2) Conclude healthy when sick (missed detection). Yup, we’re doing an ‘hypothesis test’ — can’t avoid it. Yup, can call the false alarm ‘Type I’ error and the missed detection ‘Type II’ error. So, have data, say, on each of 30 variables 1000 times a second. Hmm …! I thought about this once: Borrow a little from ergodic theory. Get a finite group of measure-preserving transformations. Make use of a classic result of S. Ulam (best known for the Teller-Ulam configuration) on ‘tightness’.One of the problems that motivated my work: There was a ‘cluster’ of servers. ‘Load balancing’ sent each incoming piece of work to the least busy server. One server got a little ‘sick’, was throwing all its work into the ‘bit bucket’, was not very ‘busy’, was getting nearly all the work and essentially ruining the whole cluster. So, how to automate the monitoring to detect such things, quickly, in real time, with selectable, low false alarm rate and, for that false alarm rate, the highest detection rate possible from any means of processing the available data on several variables? Did that!Apparently recently there was a similar problem at a cluster of e-mail servers at Google. Generally ‘load balancing’ with a cluster of servers can bring some surprising new ‘modes’ of failure!BTW, sending the next piece of work to the least busy server is not always the way to get the best performance; a lot has been done on this problem, e.g., by considering the problem as a case of stochastic optimal control!For the diagnosis, some work has been done on this, also. Start with ‘root cause analysis’ which looks for a single cause that can generate all the symptoms, error messages, etc. Uh, one failing node can put out error messages (say, via SNMP) but, due to its failure, result in other nodes also putting out error messages. So, can get messages from many nodes and not just the one node with the ’cause’. So, given all the messages, look for the ‘root cause’.These subjects are not just ‘craft’; there’s some applied mathematics, computer engineering, etc. here.So, first-cut, we will want to be clear on what we have as a ‘network of queues’ (for our site we have that already) and put in some instrumentation and means to collect the data in real time. Again, how much of this before versus after going live?But, DO want a lot in place before we have a ‘fail whale’ crisis like Twitter did! I see such crises coming like a high speed train just 300 yards away on a track.

9. Geeky Peek is a good one from RRE’s portfolio (http://www.geekypeek.com/). It’s jointly written by a few of the engineers at Peek (www.getpeek.com).

10. In the run-up to Windows 7, Steven Sinofsky allowed engineers from the Win7 team to post on the aptly-named *Engineering Windows 7 Blog*. Although it’s been dormant since Windows 7’s release, it was a great example of “opening up the Kimono” by looking at the messy innards of a large software engineering project.http://blogs.msdn.com/e7/

11. The Sticky post sort of question:I’m looking for my ideal blog. The art student who is transfering over into Coding. people always come into all sorts of places at different points in life. So that person must exist.I find a lot of people are talking to other engineers. Where are the people who can cross talk across disciplines? especially when that person is a beginner, so you can build mad skillz quickly.Just a question I’ve been having (frustratingly unsolved) for the past 6-8 weeks…

    1. I don’t know of a great resource for this either…but I do know there is a large under-served audience for this as people ask me for suggestions like this all the time…

    2. What would you hope to find on that blog? Technical tips? Storytelling?I got hired as a marketing writer in a startup because I could string sentences together, I’d taken some programming courses, and I could talk to the guys who were staying up all night coding their brains out.One of my favorite images was the guy in his white socks (no shoes), eating a bowl of Fruit Loops and milk at his desk while he described to me the functions of the code he was working on at the time. It was a key piece that would contribute to the product’s “sexiness” at the trade show the next month. I needed to know, because I needed to translate his enthusiasm into a white paper.There’s nothing like being right in the environment to pick up mad skillz.

       1. How to teach yourself without getting moody. Both storytelling andtechnical tips. Or technical tips through storytelling, would be my bestguess. And code that isn’t just the bulky I build something that ended up acompany. All sorts of code instead. In all sorts of languages about allsorts of issues in code… Like I want to see someone who decided to codeup microcontrolers for his room as a way of learning to code, and anotherperson who did it for art, and someone else. At different stages of life. And how they hacked it. And how the problem was solved. Without soundinglike a textbook. Almost like the happy club of yes this a problem, but yetyou can figure this out.I have this weird idea that despite the image of “oh I’ve been coding sinceI was 9” the vast majority of good programmers out there, probably haven’t. People come and go with skills across their lifetimes, and can hold manycareers. I’d like to deflate that image of that 9 year old who doessomething.Ok so I have a weird request. A very DIY world of code, where people comeat it from all stages and kind of teach each other what they are learning. So I can come into this blog tomorrow with “Hey I am learning to codebecause I want to make this cool art project” and this would make totalsense, and there would be a progressive simple way of me learning so withoutimpressive amounts of pain..

          1. There’s an old book from the “minicomputer” era by Tracy Kidder, called “Soul of a New Machine.” The book was beloved by both the literary and engineering crowds. You might look at that for a little of what you’re asking for.This is an interesting idea. You’ve got me musing.

             1. I loved that book. Read it in 1989 in a Wharton undergrad entrepreneurship class of all things. It changed everything for me.

             2. Glad to get people musing.

          2. Shana – your request for a blog where people come to coding from other walks of life and are learning it in order to create something sounds like the ethos of the ITP community. It is an NYU graduate program focused on working with new technology to create things (things being very broad: art, products, entertainment, social applications, etc.). The Physical Computing course teaches students how to work with computer interfaces beyond a keyboard and mouse. Here is a link to the class website: http://itp.nyu.edu/physcomp…. From there, if you are willing to hunt and peck, you will see links to exactly what you are looking for–documentation of challenges people faced with their projects and how they tackled it. Of course, you will be at a bit of disadvantage as you won’t have the context of the class to reference, but it may worth taking a look.

             1. Do you go there? I took a look- I mean while microcontrollers are reallyinteresting, for some reason I’m more attracted to the “softer more”theoretical world” of software. I mean software software. I would ignoremachines for browsers if I could (I’m realizing that this is impossible…)For a some time down the line though, I keep thinking of ITP (and a fewother programs) -however I take a more negative “Russian”/critical theory(aka lev manovich?) type view on modern internet and media when it comes toart. Throw in a little pop and a little simplicity….And I keep thinkingI’ll drive most places that actually will offer the classes I want bonkerswhen I say things like “I want to program a Spam Machine to expose ourweakness of understanding content” or “Create a room that is hooked intofacebook. What you do in the room becomes socially networked (in a creepy,big brother sort of way). Or a exhibit that showcases the strengths andweaknesses of mechanical turking. Something like that….I’m figuring if ITP or similar is an option on the table I should work on myportfolio by myself anyway for a few years. Beyond that, I’m still afraidof being the freak for liking slightly more misanthrope, critical,ambivalent messages in media. I don’t feel peppy enough for theseplaces…(which is odd, because in person I’m actually somewhat peppy)

             2. Hi Shana – I went there (graduated 2001). What is great about ITP is that they encourage people to apply from all backgrounds (mine was journalism and management consulting, I had classmates with back grounds in music, design, programming, dance, education, non-profit, film and more–and from 19 countries). You should check out the course descriptions and go to an orientation or talk to someone in admissions. ITP is what you make it, and you can make it just about anything, catered to your interests.

             3. Thanks!

             4. ITP is amazing. i highly recommend it. everyone i’ve met that went there isa great person and super talented. many of our companies were founded by orstaffed by ITP grads

12. Redfin engineers have a blog: http://blog.redfin.com/devb…Whenever we solve an interesting problem we ask ourselves whether to blog about it or keep it to ourselves, particularly since we tend to solve problems of a type (e.g. geospatial indexing in Postgres) of special interest to our competitors. We almost always decide to share, in part because a few of our engineers are also contributors to open-source projects. When engineers blog, the result may have more practical benefit for readers than essays from other authors, which is on balance a very good thing.

    1. “we almost always decide to share”me too 🙂

13. I love reading “behind the scene” blog posts on a service’s engineering. It’s so easy to take simple features for granted when you don’t know all the challenges involved. So when I see a post or a presentation revealing an inside glimpse, I’m all over it.It’s for very similar reasons that I love watching the b-roll of movie productions. It makes you feel like you’re a part of the magic.Most web services share similar problems but with varying twists. I think we have some pretty unique development challenges at Disqus given the nature of what we do. Anton’s post on the blog was great and that’s I’m trying to push for more of.

    1. Disqus is a very impressive commenting service, not too many can appreciate what goes behind the scenes, as we only see what’s commented on.

14. I work at kaChing and we have a great engineering blog at http://eng.kaching.com where we talk about solutions we’ve come up with. It’s freeing to have a company showing how it works, warts and all, and we hope others can learn from what we’ve been through.

15. The VPE at Wordnik.com (where I work) just wrote a great post on the challenges of data storage and retrieval with the very large (many billions of words) corpus that powers our site.

16. This is a much needed trend. Let’s open up.

17. Hello Fred and Shana, work in progress: http://avccommenters.blogsp…

    1. I like this resource. Thnx for putting it together.Ping me as it gets further along.

    2. that’s awesome.kid mercury will tell you that shana’s list and your list should be default parts of a community platform that this blog and all blog communities should have access to

18. Gustin Prudner of Entryway is very much focused on the craftsmanship of coding – he and I were talking about this on Friday, and you can see his blog entry on ‘Craftsmanship by Any Other Name is Craftsmanship’ from Jan 28 here:http://www.gustin.cc/This was his entry in The Wandering Book, which focuses on software craftsmanship and invites anyone to contribute their personal thoughts on what it means to be a software craftsman:http://www.nexwerk.com/the_…

    1. Thanks. I’ll check them out

19. The code itself isn’t the craft … it’s just a tool like paper, scissors, tape, or glue. It’s what you make with those tools that becomes craft. I often use/mis-use/abuse the tools to create (what I think is) beautiful art. Crappy code, elegant solution.

20. How does one reconcile sometimes contrary mandates for early-stage start-ups: – Get to the marketplace to test the user-case and business propositions, which – more often than not – lead to code hastily thrown together and shipped out the door – Building a software basis that is architectural sound and elegant, which can slow down the whole process.Naturally they don’t have to be mutually exclusive, but there is a bit of a rub. Typically, much more seasoned tech talent can’t reconcile the two goals almost subliminally and achieve both, drawing on experience of using well formed software design patterns. However, these seasoned people typically are going to be in a different set of salary requirements – generally being older and perhaps with family.Are you – as an investor – part of that debate between “doing-it-quick-and-cheap”, perhaps believing that the code can be re-architected/re-written once the business proposition has been verified, and “build-software-that-can-scale”, both in terms of technical performance, but also in terms of re-usability and maintainability?Naturally being on the hand-on tech side of things, I tend towards the latter, but I do find myself constantly having to “sell” the idea of slow-and-steady.

    1. that’s a great question. I’d love to hear the answer to that.

    2. Umm. Need to fix a critical typo, that’s supposed to be:”Typically, much more seasoned tech talent *CAN* reconcile the two goals almost subliminally and achieve both, drawing on experience of using well formed software design patterns”

21. And yet etsy buyers must face a poorly developed parametric search. So from this buyer’s perspective, their “code as craft” could use a good dose of technical science.

    1. they are working on it right now. they acquired a company late last yearwith real expertise in this area.2010/2/22 Disqus <>

       1. Hello Fred. Thank you for your reply and I’m glad to hear that etsy is working on this very key element to future success. Regards, Ardith

22. I think a Development blog is not only a great opportunity for developers to write and to show respect for their work, but it’s actually also an amazing way to drive traffic to your corporate sites. A lot of the topics that developers write about are very specific and optimized, and they can lead to a nice long tail of traffic.We broke our corporate blog into two, Marketing and Developing, last year and the marketing posts now account for over 80% of our traffic. You can see the type of things we post here: http://www.newmediacampaign…, and how those are naturally targeted for search results and certain communities.Another company that I know of that does the same is Viget w/ their VigetExtend Developers blog: http://www.viget.com/extend/Both blogs not only serve to give developers an outlet, but also as a way to drive traffic and leads to the corporate sites.

23. I work at Silentale, and our development team has a blog they update when they figure out solutions for technology that other developers might be using (beanstalkd, cacti, AWS etc): http://geeks.silentale.com/

24. Ok, so I just returned my Chi flat iron, because frankly it sucks. IT didnt get hot enough and it didnt straighten my normal slightly wavy hair. Also it felt soo cheap. Well, i was at the mall today, and this guy from the kioski was trying to sell me this straightener called the AMIKA flat iron that retailed 250 or something like that. It worked really well, and straightened great, but i noticed that the heat only went up to 200 degrees. IT got really hot when i tried it on my head, but that was a shocker because the CHI says it goes up to 370, and it wasnt even as hot as the amika. Well, anyway, the guy said i could have the iron for 100 bucks with a shine serum and a hot iron stand for free. I only paid 75 for my chi, and I didnt really want to pay more than that for its replacement, but i would if it was worth it

25. Maybe 10pcnt do. But its an aspirational mission. A big hairy audacious goal.

26. IMHO the problem with the etsy model and most similar VC-backed models is that the economic keystone (the platform, the organization that manages the whole pie) is going to take too big of a piece. to really get free agent nation, a leaner, more decentralized version of etsy is needed, IMHO.though i am not an etsy basher and think the company has its place, albeit not as a major driver of free agent nation IMHO.

27. 10%? The next paragraph in that article says that Etsy lists 50 users (out of 250,000) who make a full time living of it. It also says that Etsy puts downward pressure on crafts prices, by putting local craftswomen (e.g., at fairs) in competition with those elsewhere.Not that users should find this surprising, though — it takes a long time to knit a pair of gloves, and it’s tough to sell your own wares at a large premium if you don’t have some sort of brand or cachet. The way to make any money off of this situation is as a capitalist, skimming a little vig off of every transaction — like Etsy (or eBay) does — not as a craftswoman/laborer. I’m guessing folks who haven’t figured that out never heard of the phrase “big hairy audacious goal” because they don’t read Jim Collins.

28. ! (for KidMercury’s kick ’em while their down, boss you got dissed obligatory post)

29. And it certainly does happen: I have a friend who started selling on Etsy while home with her son. She’s now part owner of a store in Beacon, along with a few other people that I believe she met through Etsy.That actually highlights one of the other big points for Etsy (as well as developer blogs, for that matter): the online presence is in part a tool to facilitate community building. The “Etsy community” has likely built a good number of both personal and professional relationships that outlast the relationship with Etsy.

30. Prokofy would jive with this comment (where’s he been)Part of me does as well. Can’t all forms of creative contributors make a living doing what they love. If this isn’t true I want no part of crafting yet another “skim off the top system”.I believe startups and businesses have to provide value. Any community that forms around those businesses and adds value, will get at least as much back as they put in (it’s ideal motivation for value creation). Based on network effects, contributors will receive more than they deserve (connections, additional sales channels, time saved, meaning, belonging, pie).

31. I said ‘maybe’ and it was a guess. I am sure it is more than 50. And it also depends on what ‘make a living’ means. Does it mean supporting a primary wage earner while being a stay at home parent?And please don’t base all you know about etsy on a negative story about themIts a terrific company and very inspiring in so many ways

32. There is downward pressure because a lot of objects are similar, and it is hard to differentiate on quality (you can’t fully see the stuff, sorry)

33. prokofy is a woman guys.and etsy creates a ton of value.

34. I don’t understand what I’m supposed to jive with here.Etsy is a good thing. It should get more play, and there should be even more things like that, not only for crafts, but service, and not just coding and freelance editing/translation/film making which seem to be the only types of sites. It all needs more development. And it’s ok to have a capitalist skim from it if he is aggregating the eyeballs to the site.That said, I feel too often Etsy is invoked as a kind of “you girls go over there and make those nice lace doilies now, and we boys will do the more high-paying stuff like coding”.Creative Commons is of course a big destroyer of value by browbeating people into giving away everything for free. It’s actually funny to see how many zillions of people on Flikr *don’t* want you to copy their photo at all, period. The Creative Communists would have done better making the entire Internet like Etsy, always mindful of how people can get paid, and how those who are entrepreneurial can organize the marketplace for buying and selling, like ebay and amazon. But I sometimes suspect there is only room for one of each giant in each field, and these oligarchs do not make an ecology around them…

35. Somehow karmicly well fit to this post, coders, etsy, and my confusion based on the avatar.Thanks for the correction. Quiet this morning.

36. Prokofy is interesting. In person she is a woman. Online I once asked her for a gender pronoun and she answer that she was a russion, and didn’t specify a pronoun. I don’t know what she really thinks about the matter about her display of gender, but whatever she does think, she’s really interesting in her display.

37. I think Etsy’s model and Squidoo share some similarities (along with many other portals/stores App store etc.)The business brings the market, and you have to decide if that market is worth whatever cut the business takes.

38. “I am sure it is more than 50.”Not to put too fine a point on this, but there’s an orders of magnitude difference between your initial guess and the number the article got from Etsy itself — 10% versus 0.02% ( 50 / 250,000 x 100%). Checking Etsy now, it looks like the number of full time sellers is up to 101, so if the number of Etsy sellers has remained constant at 250,000 (has it?), the percentage of sellers who do this full time would seem to be about 0.04%.

39. Word

40. Great story.That community interaction provides greater value than the revenue split members receive from the gravitating site. That’s real value in my book all the way.

41. A free agent nation sounds great in theory, but in real life, most people don’t have the capital (human or financial) to be successful as free agents. You want an example of a free agent nation? If Michael Bateman is right, look at Serbia or Bangladesh. Here is Professor Bateman, from a letter he wrote to the editor of the FT (which I quoted on my old blog): [T]hanks to microfinance there has been an accelerated proliferation of informal-sector microenterprises in Serbia over 2004-08, so the country is now chock-full of traders, kiosks, shops, street-traders and subsistence farms. The base of the economy is quite simply being destroyed. […] The East Asian countries managed to develop brilliantly through channelling much, if not most, of their savings into serious growth-oriented sustainable business projects. This is why many east Asian countries may have started at similar GDP levels as Bangladesh in the 1970s but have since then massively outpaced Bangladesh in terms of growth and development. Economics 101 shows conclusively how critical savings are to development, but only if intermediated into growth- and productivity-enhancing projects. If it all goes into rickshaws, kiosks, 30 chicken farms, traders, and so on, then that country simply will not develop and sustainably reduce poverty.

42. There are two aspects to this problem kidmercury – the sharks who want a large percentage for little value add AND the people at the other end who think all software, music, videos, etc. should be free. Both sides make things worse. The VC who wants a reasonable return and a marketplace willing to pay for something that has value would turn everything around.

43. hahahahhaha!

44. “most people don’t have the capital (human or financial) to be successful as free agents.”agreed, but IMHO that will change, but i think it will take new models of investing first. models that involve investors investing little capital and more knowledge to build very low cost businesses are where we will see free agent nation really manifest fully, IMHO.

45. IMHO the people should continue demanding more and more for free, as perpetually falling prices is a good thing for society at large. the key, i think, is to find out how to make what is given away for free as low cost as possible. if microbusinesses work together to keep their shared costs down, i think it will help immensely in creating free agent nation, and i think future iterations of the etsy business model will be based on helping these microbusinesses work together and share costs to increase their profit margins.

46. Etsy takes a 3.5pcnt transaction fee. That’s really not a lot for matching a buyer and seller. I’m in total agreement with your comment. Well said

47. Now you see the advantage of symbols. Disqus makes me put a space after the ! Kid programmed response (min of 2 characters)

48. Etsy is a ripoff bro. Plain and simple.

49. I tell people to sell stuff on Etsy. Not because it is a full time job issues. But because of the following: If people craft anything, or make anything, they are going to end up with piles of stuff. And it needs to go somewhere..I figure once I get a full time job I will want to go back to nude drawing (I think it is one of the baselines of good thinking in a relaxing way for me, don’t ask). But then I am going to end up with stacks of drawings. And by stacks, I mean stacks. So I might as well open up an Etsy store, because I don’t want to keep them all.