Tag Archives: Google

Where the Net is Broken

Couple of examples from the past month on how the internet is not working like it should.

More groupthink instead diversity. Over the past year or so, I’ve been taking my funny picture collection, which has been threatening to take over my hard drive, and moving it to the net. Nothing special or fancy, just a bunch of pictures tagged up so that I (or others) can find them. So if you’re looking for a funny picture of a turtle or a poster you might could put up in your cube, you can find it easily. This is turning out to be a lot of work, so I decided once a week to do 16 and post them on my feeds. That way my friends can get a good laugh — and I can make sure I’ve spread the word as much as possible. It also makes me accountable to others so I can be motivated to get them all moved over.

I ran into this graph of site traffic, which took me a bit to figure out.

graph from Google analytics of my caption of the day site

As you can see, I keep adding stuff and people keep slowly coming by and using it. But why would traffic level off? After all, I’m continuing to add more material. And it’s the same kind of material. The problem, best I can figure, is that I have several thousand followers on Google+. Even though each week I clearly post something along the lines of “Hey guys, this is my thing. Each week I post funny pictures on Friday. Please ignore if you don’t like” Some folks don’t get the memo. When they see a picture of a cute baby that they don’t like because it’s trivial or just noisy, they click the spam button. So Google marks spam as coming both from my account and my blog site.

This effectively means if you have a lot of followers you can’t continually post things from your site that some of them might find worthless and/or annoying. Google is “helping” you conform to groupthink, i.e., basically rewarding (or not punishing) you only if you think and say the right things. Yay Google.

Empty wastelands of groups. I’m a member of an Agile group on LinkedIn. It has hundreds of members, but oddly nobody posts there. I really don’t know why. One guy last week asked a question about Agile architecture. I asked a clarifying question, hoping to draw in others, but there were no takers. There are several other LinkedIn groups I am a member of which are like this. Why have a group where nobody posts? I felt sorry for the guy. You’d think a large group on X would be the place to ask questions and start discussions around X, right?

Along those lines, I’m a member of the HackerNews Facebook group. It also has no posts — maybe one every couple of days. And it has thousands of members. So a few weeks ago I took to posting my blog entries over there. Who knows? Maybe it’ll start up some conversation. And it did — I got one guy posting that he hated people putting their blogs in the group and another guy telling me that I should strive for posting higher quality material to the list. The second guy said he almost clicked the “spam” button, but decided to comment instead. Thank you!

To me, the purpose of the “like” button is to take a large stream of things and help sort it out based on your preferences. What I was hearing (and what I a suspect is going on over at LinkedIn) is folks only wanting to post the very best material in a group. There’s a lot of self-editing. So nobody posts anything. But the system only works with lots of data. You shouldn’t be trying to write an encyclopedia. Every post shouldn’t be a special snowflake. Instead you should be letting it all hang out and letting the system do it’s job. So instead of active boards centered around user interests, we get these thousand-person groups where 2 people may post each month, and that’s considered too much.

I fully understand and support the idea of having group standards — you probably don’t want my funny pictures of cats in your Agile group. However if the standards (self-imposed or not) are such that two thousand people can be there and only one person posts each week, something is broken. At such a low volume, I don’t even see or notice any of that small volume when it finally does appear. There’s too much other stuff going on in my feeds. And if it’s going to be like that, what’s the point of even having the group?

Not trying to rant. Forum designers set out to do one thing, but then they try to avoid one kind of failure so much that they end up falling into other traps. Good to point these out. Can’t fix what you don’t see is broken.

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

Google Analytics Doldrums

I have between 20 and 30 sites on the internet. What can I say? Some folks collect shoes. I collect domain names and then make content — apps, targeted text, or mixed-mode stuff — for them. Or at least most of them.

I also use Google Analytics. Great free tool. But heck if I can figure out what’s useful to do with it aside from just a few measurements. Looks like there’s all kinds of stuff in there. And they just keep adding stuff, like the new upgrade. Let’s describe a few usage scenarios to see what I’m talking about.

It used to be under the old interface that I could see all of my web sites at one time and get a feel for how they were doing. Now I can’t even do that. So when my wife’s hamburger casserole recipes site began losing traffic, it took two months for me to figure it out. I can’t just scan 20 sites in one column and get a delta from the previous month. That’s not a good thing. You must understand that most of these sites make nothing at all, perhaps only enough to keep their domains renewed, so they’re all competing for my attention. If I can’t look at all of them at once and see their changes over time it destroys my workflow.

I like the new real-time metrics. Nothing like posting a Tiny Giant Books article about being a ScrumMaster on HN and watching hundreds of people come to visit. What great eye candy! But what to do with all of those little flashing circles and cool maps and stuff? On the real-time graph I can at least get a sense of where the traffic is coming from. Perhaps if lots of people are visiting from DZone I could go there and see if there were any questions about the content. That’s what I keep telling myself, at least. Hasn’t really worked out that way.

Paycheck Stub has also suffered a bit of a traffic hit, but only in the last week or two. I’d like to know why, but heck if I can figure out how to get GA to cough that up. I think it’s a change in search term usage by readers. Not sure. It’s the type of thing where I could spend 6 hours tracking it down only to make an extra few dollars a month from the resulting fix.

My funny picture site, Caption of the Day, looked like it was taking off for a while, but has leveled out recently. Are different people visiting it now than a few weeks ago? Do people not find the images as funny? Has somebody flagged the site for some stupid reason? I have no idea.

It seems like with web site analytics you’re always swimming in tons of raw data and somehow magically you’re supposed to make some kind of sense out of it all. Developers just keep adding more bells and whistles, continuing the user information and option overload.

It’s not that none of it is useful. I love using average time on site and bounce rate to compare how useful each site is to the readers. It’s just that there’s so much stuff, and very little of it is immediately useful as actionable data. For this blog I can tell you that there were 70 thousand visitors last month, 25% of which were returning friends, but that really doesn’t mean anything. Do those 25% visit regularly? Don’t know — aside from that statistic, I can’t find out. There’s a “frequency and recency” tab, but I’m not sure it’s answering the question I have, which is about the population, not the site.

Part of the problem here is that as a website content creator, I have questions about the people that are visiting, but the reporting is focused on the site. it’s a subtle difference, but to me it means the difference between being able to walk away with something useful to do and just kind of mindlessly staring at a bunch of numbers and graphs. How many logical segments of people are there? I would expect groups like Mac User-College Access-US or FB User-IE-middle-aged. If the system could propose a few demographic clusters like that, then I could use those and diving down into the rest of it would make sense.

I create stuff for people, I don’t publish data to a server. There’s a big difference. My questions are about people, not visiting statistics. I’m not getting results from a survey, I’m trying to make something people want. It’s like I’m always trying to translate from one system to another. Some days it feels like I’m in training to become a psychic.

I understand that there’s no magic bullet here, but there is room for improvement. If the answers are framed the wrong way, no matter how you ask the question you end up frustrated.

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

Good and Evil in the Garden of Hackerdom

I was reading today about some technology reporter-editor who is mad at another technology reporter-editor and it occurred to me how many hours I’ve wasted spending time watching pointless drama on the internet.

No, I do not care if the latest iPad has a video camera. No, I do not want to know what sorts of personnel shakeups are occurring at Microsoft. And no, I am not at all concerned about how venture capitalists choose to waste their money and time.

I understand that some folks do care. Perhaps you have to pitch some of these VCs. Perhaps you are interviewing for a job at Microsoft. Perhaps your business model heavily depends on video being available in portable formats.

But I suspect not. I suspect that articles like this attract internet readers because it gives them something to sit around on their fat assess and pontificate about. Apple getting a new CTO? Well you know, this guy was heavily invested in company X, and company X is highly regarded among this community.

WTF? Come on, guys. If done anywhere else about any other topic this would be called out plain and simple for what it is: idle gossip. It’s the technology version of painting the bike shed: somebody can have some sort of drama anywhere in the world that involves technology, and suddenly we’re all speculating on motives, emotions, and impact. Bickering with each other over semantics and versions of history.

Don’t get me wrong. I think a lot of such communication is good — it helps us form and maintain a community. But it’s critically important to recognize what these topics are: social candy. A little bit is fine now and then, but it’s very easy to do too much.

I just got an email from a person who is also doing some fine work around hackers. He has a monthly newsletter with all sort of good advice. He’d asked to include an article of mine, and I was more than happy. In his follow-up, then he asked if I had a chance to read the latest issue.

Well gee, I do not like being so blunt, but no, I have not decided to consume yet more material of a randomly technically tangentially interesting nature. I’m sure its a great publication, and I’m sure with my technical bent I would love reading it, but I’m really trying to be a bit more useful in my life.

And I think that’s the insidious nature of the problem. It’s very enjoyable to read about some legal battle involving two big corporations which we have attachments to, or hear about some programmer somewhere who made a million dollars attaching lasers to flying turtles, but it’s probably not a good use of our time. When we consume these things, it’s like we’re stuck in neutral, just ambling around without direction. That’s a fine way to be every now and then, but — like I said — it’s very easy to do too much of it.

This gets back to my application of Kant’s Categorical Imperative to technology: don’t put a constraint on a user unless you are willing to put that same constraint on every user who seeks a similar goal. It’s very likely you don’t need a login service for your web-app (unless there are some monetary or privacy reasons, of course). You don’t need to email or text me when your service upgrades. You don’t need to pop up in the corner of my window just to tell me that you’re still running.

These are all things that as programmers we have done at various times, and I think they are all evil to a certain degree. They take millions away — if only for a few seconds — from whatever they were doing before. And for no good reason. Daniel’s rule of big numbers: anything times a big number? It’s also a big number.

I just wrote this blog article. It took you ten minutes to read it. If there are six thousand of you, and your time is worth 40 dollars an hour on average, I just “consumed” forty-thousand dollars of worth in the world. Now it’s up to me to make sure I provide this value back in my content. And that’s assuming that you can immediately switch back to whatever you were doing after you finish reading this, which studies have shown is highly unlikely.

More of us should be doing these calculations.

The same goes for Technology Drama — stories that provide emotional impact, that consume large portions of our energies, with little or no return for us directly (except for advertising revenues for those who provide the drama)

It’s evil. Perhaps not evil in the way we’ve grown up understanding, but evil nonetheless.

After 9-11 I was on a political board and we were talking about the nature of evil. “Well,” one commenter said, “what is evil, anyway? It’s all so vague. I don’t think it has a definition.”

To which I replied, “Evil is somebody who wants to come over to my country and kill me — to take away my life”

Of course that gets into whether or not there is a universal definition of evil, but lets not go there.

That definition of evil was true then, and it’s true now. Evil is people creating material that is purposely designed to take away my life — if only a very small part of it — for somebody else’s benefit. Killing one person is a horrible tragedy. Aside from your religious or moral feelings, the world lost the benefit that one life could have provided. Playing out your drama about some fanboy topic for fifty-thousand people that consumed several hours discussing it? From the world’s standpoint, same amount of harm done, perhaps more.

Searching was the killer app of the 1990s and 2000s. Filtering is going to be the killer app of the 2020s and 2030s. Either that or we’re all screwed.

We hackers and programmers are a lot more involved in the world or good and evil than we’d like to admit. Or that we’re comfortable discussing.

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

My Master Plan to Destroy the Internet as we know it

What’s the minimum amount of interface that you need to do the most work?

I’m not talking web interface. I’m talking interface period. Picture a black box. What kinds of buttons, lights, or displays would you need to do, say, 80% of the work you do online?

I think the answer is very surprising: not very much. In fact, I’d argue with a few buttons that double as lights, and a mostly-plain display, you can do much of whatever you do right now. No keyboard, no web browser, no flash, no iPad.

Another way to phrase this question: What are the limits of a universal Minimum Viable Product?

It’s a magic box. A box where you push a simple button to get a category of information you are interested in, not a specific set of information from a specific site. Interested in technology news? Push the technology news button. Up pops a list of technology news gather from, well, anywhere/ Where it comes from doesn’t matter. Strangest 500-pound gorilla in the room ever — we go to this branded sites and participate in sign-ups, voting, games, and other “sticky” and “engagement” activities not because we particularly like them, but because they use human psychology against us in an attempt to own an entire category of material.

Want to chat with your friends? Push a button and say something. It goes to your friends (or “followers” as Twitter calls them). Send a person a message? Click on their name and speak the message. Want to review your financial status? Push a button to see your net worth. Another one to track recent activity.

Want to know why most people use Google? It’s not because of the quality. Most people use Google because after thousands of searches, they’ve trained themselves to think of “search” and the Google logo. I know — I’ve been using Duck, Duck, Go for the last 2 weeks and it has been painful. It’s a better search engine, yet in my mind the Google branding that I’ve subjected myself to over the years still draws me back to Google.

This is stupid. It doesn’t matter where all this information comes from. You want to send a message to a friend, do you care if she updated her personal email on LinkedIn? Does it matter? You just want to send her a message. We live in a wonderful age of computers. Why are we getting so wrapped up in channels, walled-gardens, brands, and all sorts of implementation details that have nothing to do with what’s important in our lives?

Such a device is known as a “magic box”. My current working title is “magic brick”, since the goal is to make the thing as plain, unattractive, and simple as possible. You can see an early prototype reading tech news from several different sites in the video below if you like. I’m building it in F# on a .NET stack, but the idea is to run it on mono on a stand-alone device

Of course this kind of detachment from the inner workings of the web not only brings freedom, it also threatens millions of established business models. Why build a freemium service if nobody comes to see your engagement material, nobody “plays” your site enough to become a paying member? How can you sell advertising if nobody ever reads the ads?

I love saying this next phrase because it has such a mad scientist feel to it (picture Dr. Evil with his pinky in his mouth) This will destroy the internet as we know it!

I have no idea where this project would lead, but I know the underlying technology is sound, I have a working prototype, and I know exactly how it’s all going to come together. If you’d like to be a henchman or a minion, we’re now taking applications. We’re also looking for places to build our secret lair. A death ray would be awesome, but perhaps that’s a bit much for now.

Technology has become an amusement park you have to drive to and pay to get in, even if “paying” only means putting up with ads and branding, and not like the human-empowering super-tool it was supposed to be. Let’s fix that.

UPDATE: Nice darker look at the situation from my reading today

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

Blue-Sky Startup Ideas: Socratic Linking

How many times have you visited a blog or read an article and wanted to reply or comment but didn’t because of these issues:

  • There was no facility to comment
  • You were confused and had a question and wasn’t sure anybody would respond
  • You didn’t want to go through the hassle of having to subscribe or revisit the site all the time just because you were interested in one article
  • All the existing comments were jokes or one-liners that added no value, and you had something serious to say
  • You really wanted to start a conversation about a topic that was related to the article, but a little different, and was afraid nobody else would be interested
  • There were obvious flaws in certain parts of the article, although some parts were good, and you didn’t want to have to take the time to point out which parts were good and which parts you agreed with
  • The tone of the board/article is very serious, and you thought of a great joke that nobody would appreciate, but it’d be a shame not to share it
  • The traffic on the site is so low that any contribution you make really won’t get many readers, making it worth much of your time

How about this idea as a response to these problems?

Continue reading

If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.