Tag Archives: Sergey Brin

Actively Reading: Organic Ideas for Startups

Been using Diigo as a way to annotate online texts. In this case, I was as interested in the tone as in the text itself. At the same time, I kept thinking about things which seem to be missing from Diigo.
Organic Startup Ideas

One thing I like about this text is its tone. There’s an honesty, an ingenuity that I find rare in this type of writing.

  • startup ideas
    • The background is important, in terms of the type of ideas about which we’re constructing something.
  • what do you wish someone would make for you?
    • My own itch has to do with Diigo, actually. There’s a lot I wish Diigo would make for me. I may be perceived as an annoyance, but I think my wishlist may lead to something bigger and possibly quite successful.
    • The difference between this question and the “scratch your own itch” principle seems significant, and this distinction may have some implications in terms of success: we’re already talking about others, not just running ideas in our own head.
  • what do you wish someone would make for you?
    • It’s somewhat different from the well-known “scratch your own itch” principle. In this difference might be located something significant. In a way, part of the potential for this version to lead to success comes from the fact that it’s already connected with others, instead of being about running ideas in your own mind.
  • grow organically
    • The core topic of the piece, put in a comparative context. The comparison isn’t the one people tend to make and one may argue about the examples used. But the concept of organic ideas is fascinating and inspiring.
  • you decide, from afar,
    • What we call, in anthropology, the “armchair” approach. Also known as “backbenching.” For this to work, you need to have a deep knowledge of the situation, which is part of the point in this piece. Nice that it’s not demonizing this position but putting it in context.
  • Apple
    was the first type
    • One might argue that it was a hybrid case. Although, it does sound like the very beginnings of Apple weren’t about “thinking from afar.”
  • class of users other than you
    • Since developers are part of a very specific “class” of people, this isn’t insignificant a way to phrase this.
  • They still rely on this principle today, incidentally.
    The iPhone is the phone Steve Jobs wants.
    • Apple tends to be perceived in a different light. According to many people, it’s the “textbook example” of a company where decisions are made without concerns for what people need. “Steve Jobs uses a top-down approach,” “They don’t even use focus groups,” “They don’t let me use their tools the way I want to use them.” But we’re not talking about the same distinction between top-down and bottom-up. Though “organic ideas” seem to imply that it’s a grassroots/bottom-up phenomenon, the core distinction isn’t about the origin of the ideas (from the “top,” in both cases) but on the reasoning behind these ideas.
  • We didn’t need this software ourselves.
    • Sounds partly like a disclaimer but this approach is quite common and “there’s nothing wrong with it.”
  • comparatively old
    • Age and life experience make for an interesting angle. It’s not that this strategy needs people of a specific age to work. It’s that there’s a connection between one’s experience and the way things may pan out.
  • There is no sharp line between the two types of ideas,
    • Those in the “engineering worldview” might go nuts, at this point. I can hear the claims of “hand waving.” But we’re talking about something complex, here, not a merely complicated problem.
  • Apple type
    • One thing to note in the three examples here: they’re all made by pairs of guys. Jobs and Woz, Gates and Allen, Page and Brin. In many cases, the formula might be that one guy (or gal, one wishes) comes up with ideas knowing that the other can implement them. Again, it’s about getting somebody else to build it for you, not about scratching your own itch.
  • Bill Gates was writing something he would use
    • Again, Gates may not be the most obvious example, since he’s mostly known for another approach. It’s not inaccurate to say he was solving his own problem, at the time, but it may not be that convincing as an example.
  • Larry and Sergey when they wrote the first versions of Google.
    • Although, the inception of the original ideas was academic in context. They weren’t solving a search problem or thinking about monetization. They were discovering the power of CitationRank.
  • generally preferable
    • Nicely relativistic.
  • It takes experience
    to predict what other people will want.
    • And possibly a lot more. Interesting that he doesn’t mention empirical data.
  • young founders
    • They sound like a fascinating group to observe. They do wonders when they open up to others, but they seem to have a tendency to impose their worldviews.
  • I’d encourage you to focus initially on organic ideas
    • Now, this advice sounds more like the “scratch your own itch” advocation. But there’s a key difference in that it’s stated as part of a broader process. It’s more of a “walk before you run” or “do your homework” piece of advice, not a “you can’t come up with good ideas if you just think about how people will use your tool.”
  • missing or broken
    • It can cover a lot, but it’s couched in terms of the typical “problem-solving” approach at the centre of the engineering worldview. Since we’re talking about developing tools, it makes sense. But there could be a broader version, admitting for dreams, inspiration, aspiration. Not necessarily of the “what would make you happy?” kind, although there’s a lot to be said about happiness and imagination. You’re brainstorming, here.
  • immediate answers
    • Which might imply that there’s a second step. If you keep asking yourself the same question, you may be able to get a very large number of ideas. The second step could be to prioritize them but I prefer “outlining” as a process: you shuffle things together and you group some ideas to get one which covers several. What’s common between your need for a simpler way to code on the Altair and your values? Why do you care so much about algorithms instead of human encoding?
  • You may need to stand outside yourself a bit to see brokenness
    • Ah, yes! “Taking a step back,” “distancing yourself,” “seeing the forest for the trees”… A core dimension of the ethnographic approach and the need for a back-and-forth between “inside” and “outside.” There’s a reflexive component in this “being an outsider to yourself.” It’s not only psychological, it’s a way to get into the social, which can lead to broader success if it’s indeed not just about scratching your own itch.
  • get used to it and take it for granted
    • That’s enculturation, to you. When you do things a certain way simply because “we’ve always done them that way,” you may not create these organic ideas. But it’s a fine way to do your work. Asking yourself important questions about what’s wrong with your situation works well in terms of getting new ideas. But, sometimes, you need to get some work done.
  • a Facebook
    • Yet another recontextualized example. Zuckerberg wasn’t trying to solve that specific brokenness, as far as we know. But Facebook became part of what it is when Zuck began scratching that itch.
  • organic startup ideas usually don’t
    seem like startup ideas at first
    • Which gets us to the pivotal importance of working with others. Per this article, VCs and “angel investors,” probably. But, in the case of some of cases cited, those we tend to forget, like Paul Allen, Narendra, and the Winklevosses.
  • end up making
    something of value to a lot of people
    • Trial and error, it’s an iterative process. So you must recognize errors quickly and not invest too much effort in a specific brokenness. Part of this requires maturity.
  • something
    other people dismiss as a toy
    • The passage on which Gruber focused and an interesting tidbit. Not that central, come to think of it. But it’s important to note that people’s dismissive attitude may be misled, that “toys” may hide tools, that it’s probably a good idea not to take all feedback to heart…
  • At this point, when someone comes to us with
    something that users like but that we could envision forum trolls
    dismissing as a toy, it makes us especially likely to invest.
  • the best source of organic ones
    • Especially to investors. Potentially self-serving… in a useful way.
  • they’re at the forefront of technology
    • That part I would dispute, actually. Unless we talk about a specific subgroup of young founders and a specific set of tools. Young founders tend to be oblivious to a large field in technology, including social tools.
  • they’re in a position to discover
    valuable types of fixable brokenness first
    • The focus on fixable brokenness makes sense if we’re thinking exclusively through the engineering worldview, but it’s at the centre of some failures like the Google Buzz launch.
  • you still have to work hard
    • Of the “inspiration shouldn’t make use forget perspiration” kind. Makes for a more thoughtful approach than the frequent “all you need to do…” claims.
  • I’d encourage anyone
    starting a startup to become one of its users, however unnatural it
    • Not merely an argument for dogfooding. It’s deeper than that. Googloids probably use Google tools but they didn’t actually become users. They’re beta testers with a strong background in troubleshooting. Not the best way to figure out what users really want or how the tool will ultimately fail.
  • It’s hard to compete directly with open source software
    • Open Source as competition isn’t new as a concept, but it takes time to seep in.
  • there has to be some part
    you can charge for
    • The breach through which old-school “business models” enter with little attention paid to everything else. To the extent that much of the whole piece might crumble from pressure built up by the “beancounter” worldview. Good thing he acknowledges it.

Less Than 30 Minutes


At 20:27 (EST) on Saturday, November 17, 2007, I post a blog entry on the archaic/rare French term «queruleuse» (one equivalent of “querulous”). At 20:54 (EST) of the same day, Google is already linking my main blog page as the first page containing the term “queruleuse” and as the fourth page containing the term “querulente.” At that point in time, the only other result for “queruleuse” was to a Google Book. Interestingly enough, a search in Google Book directly lists other Google Books containing that term, including different versions of the same passage. These other books do not currently show up on the main Google search for that term. And blogs containing links to this blog are now (over two hours after my «queruleuse» post) showing above the Google Book in search results.

Now, there’s nothing very extraordinary, here. The term «queruleuse» is probably not the proper version of the term. In fact, «querulente» seems a bit more common. Also, “querulous” and “querulent” both exist in English, and their definitions seem fairly similar to the concept to which «queruleuse» was supposed to refer. So, no magic, here.

But I do find it very interesting that it takes Google less than a half hour for Google to update its database to show my main page as the first result for a term which exists in its own Google Books database.

I guess the reason I find it so interesting is that I have thought a bit about SEO, Search Engine Optimization. I usually don’t care about such issues but a couple of things made me think about Google’s PageRank specifically.

One was that someone recently left a comment on this very blog (my main blog, among several), asking how long it took me to get a PageRank of 5. I don’t know the answer but it seems to me that my PageRank hasn’t varied since pretty much the beginning. I don’t use the Google Toolbar in my main browser so I don’t really know. But when I did look at the PR indicator on this blog, it seemed to be pretty much always at the midway point and I assumed it was just normal. What’s funny is that, after attending a couple Yulblog meetings more than a year ago, someone mentioned my PageRank, trying to interpret why it was so high. I checked that Yulblogger’s blog recently and it has a PR of 6, IIRC. Maybe even 7. (Pretty much an A-List blogger, IMHO.)

The other thing which made me think about PageRank is a discussion about it on a recent episode of the This Week in Tech (TWiT) “netcast” (or “podcast,” as everybody else would call it). On that episode, Chaos Manor author Jerry Pournelle mused about PageRank and its inability to provide a true measure of just about anything. Though most people would agree that PageRank is a less than ideal measure for popularity, influence, or even relevance, Pournelle’s point was made more strongly than “consensus opinion among bloggers.” I tend to agree with Pournelle. 😉

Of course, some people probably think that I’m a sore loser and that the reason I make claims about the irrelevance of PageRank is that I’d like to get higher in a blogosphere’s hierarchy. But, honestly, I had no idea that PR5 might be a decent rank until this commenter asked me about. Even when the aforementioned Yulblogger talked about it, I didn’t understand that it was supposed to be a rather significant number. I just thought this blogger was teasing (despite not being a teaser).

Answering the commenter’s question as to when my PR reached 5, I talked about the rarity of my name. Basically, I can always rely on my name being available on almost any service. Things might change if a distant cousin gets really famous really soon, of course… ;-) In fact, I’m wondering if talking about this on my blog might push someone to use my name for some service just to tease/annoy me. I guess there could even be more serious consequences. But, in the meantime, I’m having fun with my name’s rarity. And I’m assuming this rarity is a factor in my PageRank.

Problem is, this isn’t my only blog with my name in the domain. One of the others is on Google’s very own Blogger platform. So I’m guessing other factors contribute to this (my main) blog’s PageRank.

One factor is likely to be my absurdly long list of categories. Reason for this long list is that I was originally using them as tags, linked to Technorati tags. Actually, I recently shortened this list significantly by transforming many categories into tags. It’s funny that the PageRank-interested commenter replied to this very same post about categories and tags since I was then positing that the modification to my categories list would decrease the number of visits to this blog. Though it’s hard for me to assess an actual causal link, I do get significantly less visits since that time. And I probably do get a few more comments than before (which is exactly what I wanted). AFAICT, WordPress.com tags still work as Technorati tags so I have no idea how the change could have had an impact. Come to think of it, the impact probably is spurious.

A related factor is my absurdly long blogroll. I don’t “do it on purpose,” I just add pretty much any blog I come across. In fact, I’ve been adding most blogs authored by MyBlogLog visitors to this blog (those you see on the right, here). Kind of as a courtesy to them for having visited my blog. And I do the same thing with blogs managed by people who comment on this blog. I even do it with blogs by pretty much any Yulblogger I’ve come across, somehow. All of this is meant as a way to collect links to a wide diversity of blogs, using arbitrary selection criteria. Just because I can.

Actually, early on (before I grokked the concept of what a blogroll was really supposed to be), I started using the “Link This” bookmarklet to collect links whether they were to actual blogs or simply main pages. I wasn’t really using any Social Networking Service (SNS) at that point in time (though I had used some SNS several years prior) and I was thinking of these lists of people pretty much the same way many now conceive of SNS. Nowadays, I use Facebook as my main SNS (though I have accounts on other SNS, including MySpace). So this use of links/blogrolls has been superseded by actual SNS.

What has not been superseded and may in fact be another factor for my PageRank is the fact that I tend to keep links of much of the stuff I read. After looking at a wide variety of “social bookmarking systems,” I recently settled on Spurl (my Spurl RSS). And it’s not really that Spurl is my “favourite social bookmarking system evah.” But Spurl is the one system which fits the most in (or least disrupts) my workflow right now. In fact, I keep thinking about “social bookmarking systems” and I have lots of ideas about the ideal one. I know I’ll be posting some of these ideas someday, but many of these ideas are a bit hard to describe in writing.

At any rate, my tendency to keep links on just about anything I read might contribute to my PageRank as Google’s PageRank does measure the number of outgoing links. On the other hand, the fact that I put my Spurl feed on my main page probably doesn’t have much of an impact on my PageRank since I started doing this a while after I started this blog and I’m pretty sure my PageRank remained the same. (I’m pretty sure Google search only looks at the actual blog entries, not the complete blog site. But you never know…)

Now, another tendency I have may also be a factor. I tend to link to my own blog entries. Yeah, I know, many bloggers see this as self-serving and lame. But I do it as a matter of convenience and “thought management.” It helps me situate some of my “streams of thought” and I like the idea of backtracking my blog entries. Actually, it’s all part of a series of habits after I started blogging, 2.5 years ago. And since I basically blog for fun, I don’t really care if people think my habits are lame.

Sheesh! All this for a silly integer about which I tend not to think. But I do enjoy thinking about what brings people to specific blogs. I don’t see blog statistics on any of my other blogs and I get few enough comments or trackbacks to not get much data on other factors. So it’s not like I can use my blogs as a basis for a quantitative study of “blog influence” or “search engine relevance.”

One dimension which would interesting to explore, in relation to PageRank, is the network of citations in academic texts. We all know that Brin and Page got their PageRank idea from the academic world and the academic world is currently looking at PageRank-like measures of “citation impact” (“CitationRank” would be a cool name). I tend to care very little about the quantitative evaluation of even “citation impact” in academia, but I really am intrigued by the network analysis of citations between academic references. One fun thing there is that there seems to be a high clustering coefficient among academic papers in some research fields. In some cases, the coefficient itself could reveal something interesting but the very concept of “academic small worlds” may be important to consider. Especially since these “worlds” might integrate as apparently-coherent (and consistent) worldviews.

Groupthink, anyone? 😉