Rulerbreaker

Know what you like and tell your team

Thu, 03 Jan 2013 09:00:43 -0800

We recently hired our third designer at @carezoneteam, Scott, and to get him into the team and get started working together we did something I have not done before but which turned out to work really well. We spend some structured time discussing what we find beautiful and well designed. Personally rather than professionally.

Walter, Brittany, Scott, and I started with pepper grinders and then moved on to houses, guitars, and all kinds of other products. Take for instance the Oxo measuring cup, is it a brilliant new solution to an age old problem or is it a break with tradition that messes with what a measuring cup should be? What makes a guitar? The shape, the materials, the sound? Is the Parker Fly a well designed guitar despite being made from composite materials rather than wood? And the Steinberger, Obviously still a guitar but has it lost its beauty or maybe surpassed more traditional guitars? If you lived in a house from early 1900 would you decorate it as the appropriate for that time period or would you tear everything out and modernize it? Can a modern kitchen be beautiful in an old house? Do you like minimalism? What about change? Are historical things inherently good?

You know this and we know this: we are not designing for ourselves. But as designers we are interpreting data and creating solutions and regardless of how hard we try we cannot ever be objective, our point of view will always play into how we solve problems. That is in my book an asset and not a liability, it’s what makes my designs different from yours. But if you have to work together it helps to know where everyone else in the room is coming from, it helps appreciate their interpretations and solutions.

I love the design of the Oxo measuring cup. It does not require any new knowledge or any new behaviour. It just solves a problem better than any of the previous thousands of solutions for that challenge and it shows that regardless how much people have worked on a problem there is usually still room for improvement. That is why month view in the carezone calendar let you see two full weeks of the next month. That way you can always see three weeks out into the future, something that most other calendars I have seen does not let you do. I will always try to look for a better solution even to problems that have standard solutions like signing in, form validation, etc. This means I will often introduce something new. That is not always the right thing to do but now that the people I work with know where I am coming from and understand what I value they will be able to help me see when the change is not a good idea rather than think I am introducing change for the sake of change. And hopefully also appreciate when the change really is needed and good.

Get it out in the open

Wed, 02 Jan 2013 08:54:49 -0800

Whenever we start a project I like to spend the first session with the team listing our solution ideas. Before there is data, before we quite know what we are actually designing, I like to do a little brainstorm with the team to get all our preconceived ideas for solutions out in the open and up on the walls.

I find that if we spend a little time up front getting all of our ideas off our chest and keep it around for everyone to see everyone gets more open to everyone elses ideas and not least to the data as we learn more about the people we are designing for and their needs.

We all have ideas, products we really like, certain designs we really dislike, and some vague idea of what the right solution for the task at hand might be. We are experienced professionals so of course we have a library of solutions ready to go.

The idea here is that when your idea is out in the open and up on the wall it is safe. You feel heard and if you want to get back to the idea you can easily do that, just point at the wall.

I find that when we do a little ‘purging’ up front we often do not return to those original ideas. They are usually just reflex solutions and not really appropriate for the task at hand. On the other hand I began this process after having been through a couple of projects where we kept returning the original first ideas even if they were not the right ones.

Cold brewed coffee in sparkling water. Update: Let it sit for 12...

Sat, 22 Dec 2012 00:01:00 -0800

Cold brewed coffee in sparkling water.

Update: Let it sit for 12 hours. Tasted the extract and there was a faint hint of bubbles. Made some sparkling water and boiled it. Poured a cup of 1/4th extract and 3/4 almost boiling sparkling water. Sparkles in the pour but none left to taste so the verdict is don’t bother. It tastes good but no different form regular cold brewed coffee.

The hippest blog

Fri, 21 Dec 2012 23:50:00 -0800

You can’t be cult without a cult following and you cannot be little known unless you are at least known a little.

If you write for only yourself this has no bearing. You can write a diary and be done, and if you write to get famous, get invited to do keynote talks at all the hip conferences, and get laid like a rock’n’rolla then you are better off not reading any further, but if your innermost dream is to write and accidentally be discovered, dug out of obscurity, and hailed for your undiscovered genius then you ought to pick your publishing platform with care.

May I suggest you stay off wordpress, blogger, tumblr, posterous, and any other blogging platform. I don’t even have to mention the FTG+ triumvirat, do I? Instead, use google docs. Write your posts there, change the permissions to public, and enjoy your freedom. If anybody finds anything you have written it’s their own damn fault for looking. You have published your thoughts but since no one is seeing your posts in some RSS feed or in some cacophony of tags you are free to write whatever you like. No self censuring to curate your own brand. Just you and the remains of a blank canvas, figuratively left on the table as you walk away, to be picked up by the whirlwind of the internet search engines and they alone will determine its notoriety.

Vacation. Time for food experiments. Sour patch kids in apple...

Thu, 20 Dec 2012 23:13:25 -0800

Vacation. Time for food experiments. Sour patch kids in apple juice for sour foam a la modernist cuisine. And in grand mariner. Not quite sure for what yet. Probably something flambé

Here the other day Peyman, one of our folks at @CareZoneTeam,...

Fri, 16 Nov 2012 09:00:42 -0800

Here the other day Peyman, one of our folks at @CareZoneTeam, said something to the effect of “we need a process, I am not a process person, I really am not, but we need a process.” That got me thinking, because most of us tell ourselves that we are not process people but on the other hand we like some amount of order and predictability. It also got me wondering why so called process people have such a bad reputation.

I think there is another axis at play; how rigid do you adhere to processes? Processes are rules and you can debate how many rules you want but you can also debate how strictly you want to adhere to those rules. Some people would say that you should not have rules unless you follow them and others would say that rules are just guidelines.

Personally, I am in the 4th quadrant. I like processes. I especially like them written down so we as a team can have a common language, but I am pretty lenient in how we follow them. Hope this helps you think about the good and bad of being a so-called process person :-)

3 Phone Screens

Thu, 18 Oct 2012 11:56:22 -0700

I screwed up. Majorly. I had just started at CareZone, posted a job for a visual designer, gotten a resume that was a lot better than the previous 10 and fell in love. I called up the candidate and he sounded good so I scheduled a full day of interview. In my head I had already hired him. I had already found a visual designer after two weeks and for someone who is afraid of failing and love to exceed expectations this sounded great. Alas it was not to be. The day came. I managed it poorly and the only good thing I did that day was man up and cut the day short after lunch. I felt terrible but I am sure I didn’t feel as bad as the candidate.

I did a poor phone screen and only I did a phone screen. We needed to change the process so now we make three phone screens. Actually we do more than that. Chi-Kai our product manager takes a first pass at candidates and whomever he thinks is worthwhile comes over to me. I schedule phone screens with the people who seem promising and if that goes well we do two additional phone screen. One with Brittany, our visual designer, and one with someone else, typically Kristina or Chi-Kai to get a broader sense of the candidate. For skills and company fit.

At this point we have winnowed the amount of people down quite a bit. And then it’s on to the real interview. Another post, another day.

It may seem excessive to do three phone screens but I think it makes sense. At Microsoft one of the Vice Presidents (manager of about 1200 people at the time) asked his 5 General Managers to break down their recruiting funnel. Something stood out. One group only hired 1 out of 12 people they did full day interviews with and another hired 1 out of 3. The conclusion was that the latter group was not critical enough. That is until they looked into how many phone screens each group did. The latter did two to three phone screens per candidate while the former only did one. The group that only hired 1/12 interviewees were simply sloppy and wasting their own time, money, and not least the candidates’ time.

I re-learned that lesson quickly at CareZone so now we do more phone screens. And yes, it’s time consuming but a whole lot less than flying a candidate in for a full day of working together only to find out that it was not meant to be.

Another processing experiment. Using only red and squares...

Wed, 19 Sep 2012 09:07:00 -0700

Another processing experiment. Using only red and squares instead of circles. Looking for the effect on three different font types.

Thoughts on Hiring

Tue, 18 Sep 2012 08:56:50 -0700

Hire someone that scares you

When you are recruiting you get to read a lot of resumes and you probably end up browsing LinkedIn quite a bit and you see some impressive portfolios and read some impressive resumes. It can be quite intimidating. You can’t help but think about what will happen to you if you bring in these amazing people who are clearly so much better than you. I believe that is healthy and normal and that those are the people you want to bring in. These are the people who will stretch you and make you strive to be better.

I hired Brittany, our visual designer at CareZone because she, through her work, made me see how much I had to learn about visual design. More importantly though she made me think. She forced me to look at our product differently and she showed me how we could take the product in a different direction. I am currently reading up on graphic design and practicing. I have a lot of catching up to do.

Don’t hire a copy of yourself

Bill Buxton usually says that he never hires anyone who thinks and acts like him because he already has someone playing that role. This is another way to say you should probably not hire someone who scares you because they can do exactly the same things that you can do. Unless of course you are trying to find your replacement.

This is obvious but hard because after all you are really well suited to judge the kind of skills you bring to the table and you probably also value that skillset. It can be surprisingly hard to hire someone not like yourself.

Unless it’s yes, it’s no

Everyone I have talked with about recruiting have said that unless it’s a clear “yes” it is a “no”. That can be really hard sometimes because occasionally you find really good candidates that you wish you could hire but just aren’t right for you at this point in time.

Ultimately it’s about what conversations you want to have. Everyone brings something to the table and will need help with other things. Everyone will see something you aren’t seeing yourself so you have to figure out what precisely they are going to see. I could hire a very technical designer and get some really good insight about scalability and responsive designs. And I am currently lacking that conversation but only because we do not have a frontend dev. Once we hire a frontend dev I expect that person to bring up those perspectives and then I would no longer need the technical designer. Or more precisely, with limited resources I would rather hire someone else that brings a different perspective because otherwise I will lack some other important conversations all together.

Design with Code

Mon, 17 Sep 2012 08:56:43 -0700

I am reading Graphic Design -The New Basics and got introduced to Processing. Processing is a programming language made for graphic designers. It is meant to make it really easy to explore graphic designs that change over time or are based on math or random elements. And it seems to be part of every graphic design curriculum nowadays. When I went to school something similar was just appearing but it had not taken off the way processing has. If you have programmed before it is pretty straightforward to get started. The tutorials are good and the documentation for the libraries are good. Well worth trying.

I wanted to experiment with some t-shirt designs and wanted them to be more random/organic than what I have the patience to do in illustrator so I played around with writing some black text on white and then wrote a small script for pointilizing that image and make the black text into colored bubbles. I use the word “wrote” loosely here. The final script is 44 lines of code of which all but 4 or 5 lines were either copied or trivial :)

I let the program run for several hours and saved a frame every 4 minutes or so. Here is a little movie of what that turned into. I don’t know if I want to make any of these into a t-shirt but I think it’s a good beginning to making designs a little more interesting.

Here is the code

PImage img;
int pointillize = 16;

void setup() {
  size(740,280);
  img = loadImage("529.jpg");
  background(0);
  smooth();
}

void draw() {
  // Pick a random point
  int x = int(random(img.width));
  int y = int(random(img.height));
  int loc = x + y*img.width;
  
  //add some spice
  float rspice = random(255);
  float gspice = random(255);
  float bspice = random(255);
  
  
  // Look up the RGB color in the source image
  loadPixels();
  float r = red(img.pixels[loc]) + rspice;
  float g = green(img.pixels[loc]) + gspice;
  float b = blue(img.pixels[loc]) + bspice;
  noStroke();
  
  // Draw an ellipse at that location with that color
  fill(r,g,b,100);
  ellipse(x,y,pointillize,pointillize);
}

Making Standup Work

Thu, 23 Aug 2012 09:47:39 -0700

Yesterday Jonathan got an email about gamification in the workplace. A seminar about how to use game design techniques to make his employees more engaged and productive.

A brilliant idea. Who does not want to be more engaged and productive? Time to make some experiments at work.

We have a daily standup meeting for everyone. We try to do it in 15 minuttes. We are currently 14 people and we usually start a few minuttes past 11 and often finish before 11.15 but not always. Sometimes we go to 11.20. We are in two main locations but with Trent working out of Wisconsin and several people working form home once or twice a week we usually have four “offices” on a call. Amble opportunity for fun.

Here are some rules to make standup even better.

30 points to the office that starts the meeting. Points are shared among the people in the Office that day
Everyone: Minus 1 point for the first minute we start late, 2 for the second and so on. So if we start 11.03 everyone is deducted 6 points.
Everyone plus 2 points for every minute we stop before 11.15. If we stop 11.13 everyone gets 4 points.
3 points for going first.
4 points for going last.

Additionally points are personal and carried over day to day. However they have a half life of 1 day so if you get 10 points on Monday, you start Tuesday with 10 points. And if you get 0 points on Tuesday you start Wednesday with 5 points. If you got 7 points on Tuesday you would start Wednesday with 7 + 5 points.

You can also spend points. The rule in standup is that once you have said your piece you say the name of the person who should go next. If you really want to go next, you can spend 10 points to take the turn away. So if Jonathan hands it over to Trent but I really want to go, I can say “10 points, I’ll take the turn” and then I go and then decide whom to turn it over to afterwards. Of course Jonathan or Trent can take it back from me. Either can counter and say “12 points back” And if they both say that each only pay 7 points while I still pay the 10 points.

I would love more ideas for how to spend points.

There is a service for that

Mon, 06 Aug 2012 09:08:15 -0700

One of the most striking things when going from one of the biggest tech companies in the world to a very small one is that at Microsoft we wrote all our software and in CareZone we are using a plethora of services. We use an expense reporting service (expensify), obviously a payroll and so on service, a chat service (campfire), a product management service (trello), and we use Google docs. This means you end up with a bazillion new logins and passwords and not least a long list of bookmarks of your most important stuff.

For me personally this means that my launcher is not the dock (Mac equivalent of Win7 Taskbar) but instead bookmarks in a browser.

In Windows 7 I would put these shortcuts on the taskbar and in Windows 8 I would put them in the start menu but I haven’t (yet) figured out a way to put a bookmark on the dock or on the Launchpad in Mountain Lion.

Finally it means I have to remember a multitude of logins and passwords. ALthough some of these services do let me log in through Google. That at least is good.

So far I think the services are on par with internal Microsoft services. Some are better, others worse. Nothing so far is really magic. But Trello is definitely worth trying out. We use it pretty much like UserVoice does (see their blog post).

One week. Three lessons learned.

Thu, 02 Aug 2012 23:48:34 -0700

It’s been a week now. A full week since I joined CareZone after ten years with Microsoft. A week might not be a lot of time but I think I have already learned a number of things which would have been helpful in my old job.

To set the context I used to work with people in India and China, 12.5 and 9 timezones away respectively. Also I used to work in a big design team that reported up through a different structure than the product teams I worked with. The main implication for this post was that I had more than one core team, a design team, and a product team.

In contrast CareZone is about 15 people. Some in San Francisco, some in Seattle, and one in Wisconsin. In other words we might be distributed but we are practically all in the same TimeZone (2 hours don’t really count) and our team is small enough that everyone knows everyone.

Get a persistent multi-party IM

This is one of the most striking lessons learned. At Microsoft we had IM, Skype, and Yammer. But IM was for one-off 1-1 conversations. Yammer, modeled on Facebook and Twitter, was meant as the informal thing to keep us in the loop with each other. It never really worked. It never really worked because the groups were too big. Facebook and Twitter are good for following along in other people’s lives and for broadcasting. For a small team however, you need a sense of intimacy and camaraderie. A persistent multi-person IM is good for that. At CareZone we use Campfire. It has the basic functionality you would want plus some quikiness in the form of a bot that be used as a conversation starter.

If I went back to Microsoft tomorrow I would set up a chat room for my product team and one for my immediate design team. I might also set one up for the bigger design team.

15 minutes daily standups at 11am

I have been in countless standups at Microsoft but they have never been as interesting as I think they are at CareZone. Two things I think work better than what I had previously experienced. They are shorter. 15 minutes instead of 30 and they are at 11am rather than around 9 or 10. This means everyone can join even if they are late risers and they are short enough that they do not get tiring.

Celebrate work each week

Every Monday we go over what we put into production last week. We see a little demo and we cheer. That’s cool. And it’s part of building camaraderie. And it does not have to be only code. It’s really about celebrating progress by celebrating closed work items. This also means celebrating written blog posts or other concrete measures of progress.

It might take some work to adapt this for a place like Microsoft with the long projects but I think it could work. You could have work items for specs or designs and “launch” them.

The cost of fun: 2.04 seconds

Tue, 15 May 2012 21:38:34 -0700

Saturday I went to Duthie again. This time to ride Gravy Train, an intermediate jump line with 15 or so doubles. It’s hard to get just the rhythm just right on this line. You can easily case jumps or over-shoot them. The trick is to get just the right speed so you get the backside of every jump.

This line was not in Strava so my first run down was setting up the line in my Garmin and carrying a Freelap pole for the finish. Then I did five so runs and called it a day. I know that sounds whimpy but I woke up that morning feeling terrible. Lost my voice and a major cold. Probably shouldn’t be riding but I was so decided not to go overboard.

My runs as timed by Freelap were n/a, 0.50.70, 0.51.25, 0.50.71, and 0.52.75. 2^nd and 4^th run were good. Lost a pedal in the 3^rd run and for the 5^th run I decided to not try to race it but just see what it my time would be if I just had fun and enjoyed the jumps. Sometimes relaxing is all it takes to go faster. Not so this time.

Now I did have some trouble setting up the Freelap. I missed the end of my first lap because the Freelap watch did not pick up the signal from the pole when I passed it. And I had to repeat the beginning of the 2^nd lap a couple of times until I got the stat pole in the right position.

Imagine my surprise when I got a call from Christopher, the US distributor of Freelap, Sunday afternoon. He had read my previous post and wanted to talk to me about how to setup the Freelap so you are more likely to get all your laps timed.

I had put the poles vertical but Christopher told me to put the poles horizontally and perpendicular to the trail. That way the watch is more likely to pick up the magnetic field.

That is probably why I missed out on a couple of laps the first time I used Freelap and why I had some trouble this second time. Pretty neat to get personal instructions J

And with that we can conclude that timing your runs is fun and that you go faster when you push yourself and just cruising a run will be slower than racing it.

Yup. Going as fast as you can is faster than cruising. That was going to be the end of this post. Le grand finale. The big point. That is until I went riding with Simon Lawton from Fluidride Tuesday evening and told him the story and his reaction was “isn’t it cool how fast cruising is?” I had to ask him to repeat and then explain that to me. And his point was basically the opposite of mine. Two seconds is not a lot considering the difference of effort I put in when trying to go fast.

This is what’s really inside unicorns.

Fri, 11 May 2012 20:21:27 -0700

This is what’s really inside unicorns.

Freelap vs. Garmin

Wed, 09 May 2012 14:50:18 -0700

Whenever I have talked about getting a Freelap system people have invariably wondered out aloud whether that was needed and thought that perhaps you could accomplish the same with a Garmin Edge device. Well, I got a Freelap system from www.freelapusa.com and put it to a test against a Garmin Edge 705.

The Garmin Edge can set a lap point and you can set it to auto-start. This way you can start at the same place every time from standstill and it should stop the timer at the same spot on every run. The Freelap works differently. You setup a pole at the beginning of the race and one at the end. Both send out a magnetic field and you wear a watch that is activated when you pass the magnetic field. In other words the timer starts when you pass the first pole and stops when you pass the second pole.

To be fair the Garmin and the Freelap were made for different purposes. The Garmin has a map, a darn good GPS receiver, and tracks all kinds of stuff like cadence, heart rate, and on and on. The Freelap on the other hand does one thing only. It starts and stops a timer.

A lot of us already own a Garmin or something like it so it’s tempting to just use that. As always it comes down to what your goal actually is. In short you will be trading convenience for precision. The Freelap is more work. You need to set up the poles and collect them at the end of the day. With the Garmin on the other hand you can just ride around and if you upload your data to Strava.com you don’t even have to set lap points. Strava extracts that information for you and assuming the trail exists as a segment on Strava it will just show up automatically. And that is addicting.

But you are not getting precision with the Garmin. Not the way you do with a Freelap system. And precision can be addicting too.

To test the two systems against each other I went to Duthie hill and rode 2Hi a number of times. I went to the bottom first and set a lap point on the Garmin and a pole for the Freelap. Then I went to the top and setup the other pole. Since the Freelap watch starts as soon as you are within reach of the first pole you cannot start there so I had to start a couple of yards behind the pole. That meant that even if the systems were equally good the Garmin time would always be a little longer. The Garmin would start as soon as I began rolling and the Freelap would not start until I passed the pole. Still we can compare the time within systems and get some sense of how the two compare.

You’ll note that three of the laps are missing from the Freelap. I am not 100% sure why but I think it was because I would reset the watch at the bottom of the run and sometimes I spent a little extra time getting ready before taking the next run and the watch reverted from ‘waiting for signal’ to regular clock and I didn’t notice it. Oh well. It was the first time I used the system and I figure it makes sense because the watch must use more battery while waiting for a signal.

If you log into Strava you can see that according to his Garmin Lars Sternberg of Transition bikes ride 2Hi in 0.31. It takes me a wee bit longer. But if you compare my Freelap times you can also see that I am pretty consistent. Consistent enough that without the hundredths of a second it is hard to see any progress or difference between close runs. You should also note that Garmin mostly reports 0.42 and that covers a difference of just shy of a second (0.39.79-0.40.76) as measured by Freelap.

So what matters? The instant feedback that this run was ever so slightly faster than the last run or the online competition on Strava? The convenience of just riding around or the precision that requires you to set up poles? That is for you to decide. Now that I have both systems I will use both. I find the instant feedback gratifying and once I persuade friends to buy the watch we can compete against each other on the trails. And if you really care about getting faster Freelap is the better system. But there is still a lot to be said for just riding around with your Garmin.

Occupy, are you responsible?

Wed, 02 May 2012 21:33:16 -0700

That’s it. We are over with and done. #Occupy, I watched from the side not participating but cheering silently on as you sprung up last year and I was sad that you fizzled out till now and a little excited, almost giddy that you were going to stage a comeback on May 1^st. I had high hopes that you were going to be influential. Like in 1968 and unlike the punks in 1980’ies who didn’t do much to change society but did manage to smash quite a few McDonalds. Maybe you would be the counter-tea party. Because you are right. At the heart of it you are right. There is something rotten in America.

And then you put on a bunch of peaceful demonstrations and let in blackbloc on your day and they vandalize buildings and ruin your reputation.

And you just lost all my sympathy.

Maybe that is not fair. Maybe you did not do the violence and vandalism but like any other organization you have a brand to take care of. This day was your and it was your responsibility that something like this did not happen. This is like Apple having to clean up the mess from their suppliers. They didn’t have a choice. They couldn’t say it wasn’t their fault. None of us would have accepted that excuse.

Maybe it was done by undercover cops, agent provocateurs, maybe not. Maybe it was just done by vandals. Either way you have a choice to make. Do you care about your cause? If so you prove that it was agent provocateurs or you help track down and bring to justice the vandals.

You have more video footage of May 1^st than ever before. Welcome to the future. You should be able to find faces in that media and publish the images and if they are cops it will surface. And you will have won. You will get my sympathy, my support , and my money. You will be heroes. Underdog protagonists.

If it turns out the vandals are not cops but just, well, thugs, and you expose them, then you have proven that you are responsible and worth having a dialog with. That you are serious about your cause. And you will get my sympathy, my support , and my money. This way too you can be heroes and underdog protagonists.

But if you do nothing or fail to prove that it really was undercover cops, then you have lost. My respect and any future you might have had in politics.

I sincerely hope you redeem yourself. We need you. Well maybe not you, but we need someone to say what you are saying. Whether we need you is right now, up to you.

Ps. This makes me somewhat hopeful: http://occupr.tumblr.com/post/22304956511/occupy-seattle-and-black-bloc-destruction

Photo

Sat, 03 Dec 2011 23:04:13 -0800

vizualize: Funny GUI design I made as a part of an iPhone5...

Mon, 26 Sep 2011 23:08:36 -0700

vizualize:

Funny GUI design I made as a part of an iPhone5 concept

I did with my friend Gianluca Antonini

I think this is hysterically funny. Windows Phone brought to the iPhone without live tiles.

Jam Jar Jail Song by Therapy? I have had the chorus stuck in my...

Sat, 10 Sep 2011 22:04:00 -0700

Jam Jar Jail

Song by Therapy?

I have had the chorus stuck in my head for a couple of days.