Now that I am unemployed, off to the health exchange for me! Actually no, I am on Cobra for the next 18 months, but it is super expensive. Any opinions on the msft alumni plan?

As of midnight tonight, I am a free agent for the first time in a long time. I am leaving Ignition and beginning to plan the next phase of my professional life. It has been a while coming.

I’ve been associated with Ignition for 13+ years – the longest involvement I’ve ever had with a company. The early days were exciting, I was fully engaged. But in 2005, some personal issues caused me to step aside for a while. I eventually was able to recommit time to Ignition, but I never really had the same connection after returning. I found myself frustrated with the venture model, uninspired about the nature of the work, and too often thinking that I was on the wrong side of the table – my empathy was with sweat equity, not financial equity.

And so I’ve made no new investments the last couple years, I’m not part of the latest fund, I’m not currently on a board seat. I’ve effectively voted with my feet, I don’t really want to be involved with the venture business anymore. So effective now, I am completely out of Ignition, I am not involved with management of Ignition or the funds at all. I still have an investor’s interest in the funds, but I am just another LP now. It’s an amicable separation, I wish my remaining friends at Ignition all the best, I will still help them out when asked, as I will my friends at portfolio companies.

Going forward, I am not going to be sitting in the park feeding pigeons. I am as excited about technology as I’ve ever been. I am working hard on retraining myself to be a modern developer, becoming conversant with all the current technologies and tools. I am very excited about the coming wave of ever cheaper devices and the Internet of Things (terrible name tho). It still feels like early days in the digital revolution. I am meeting folks, talking about ideas, prototyping little things, learning open source, brainstorming, etc. I have 25 devices and sensors spread all over my desk. I’m working with some other folks on all this and having a blast. We may form a company. We may do angel investing. We may do something else. At some point I, and we, will be ready to talk about what we are doing next.

For now tho, I am enjoying the freedom of a completely blank slate, and the chance to really dig in to technology. And the opportunity to do more of the things that I enjoy – working with great people, betting on the long technology waves, and figuring out the thin edge of the wedge for new technologies and new products.

“Congratulations! Your first AT&T Wireless Bill is Now Available on Video.”

Who came up with this one? What was the internal discussion around this, how did they convince themselves that “video bills” were a good idea? This better not add even a penny to my monthly billing.

So I have a Docker/Ubuntu VM running in virtualbox via vagrant, and am building new docker images that i can use as containers in another virtualbox vm running coreos, and i need to copy files around between all these, and ssh between them, and manage all the variants of images that each of these needs, and manage packages across all these (and trying to avoid using different package managers between all these), and oh yes i also spin up vms at a cloud service and need to shuttle things up there and back, and also i need to build some physical install images for real hardware, and manage all the images and packages for that, and jeepers this is all confusing as hell. so many levels of containment and slightly different containers and key management and network management and i just want to blow my brains out at some point.

Feels like there is a need for a new type of IDE that manages all this crap and keeps it straight for you. A container/cloud focused IDE. The emergent wave of cloud IDEs are great for the cloud but they don’t really help me with all this other stuff.

Maybe I am doing it wrong? Maybe I should just dump virtualbox (which is kind of a buggy mess anyway) and just shift everything to cloud vms on a particular cloud service? that has some appeal, but i am thinking about building appliances and ultimately i need code running right here in boxes here, accessing real devices and inputs here.

OK now that the miserable non-conference schedule is out of the way, on to the real season. OSU is bunched up with a large number of undefeateds at the top of the polls. OSU can probably get to the BCS championship if they remain undefeated, but given weakness of schedule, they probably need help – if they end the regular season with the same record as an SEC and a PAC-12 team, I fear the Buckeyes would be odd man out. A quick glance suggests that these games and weekends could define who plays for the championship:

• Weekend of September 29: #23 Wisconsin @ #3 OSU, #6 LSU @ #9 Georgia. Certainly OSU’s biggest bump in the road, and an LSU win on the road could vault them up the standings.
• Weekend of October 5: #16 Washington @ #5 Stanford. Go Huskies!
• Weekend of October 12: #2 Oregon @ #16 Washington, #20 Florida @ #6 LSU.
• Weekend of October 19: #8 FSU @ #3 Clemson
• Weekend of October 26: PSU @ #4 OSU
• Weekend of November 9: #2 Oregon @ #5 Stanford, #6 LSU @ #1 Alabama
• Weekend of November 23: #10 Texas A&M @ #6 LSU
• Weekend of November 30: #1 Alabama @ Auburn, #3 Clemson at #12 South Carolina, #4 OSU @ #18 Michigan, #22 ND @ #5 Stanford, #8 FSU @ #20 Florida,
• Weekend of December 7: Conference Championship games. Very possibly a rematch for OSU vs Northwestern or OSU vs Michigan?

Looking forward to a great season!

Fitzgerald “disappointed” – OK when you read the details, Fitzgerald actually is pretty reasonable, but still his word choice “disappointment” is what carries the day. And he should know that. A guy in position of power, making millions of dollars, expressing any disapproval of a young player who is under-compensated and has few rights – that is a dick move. Saying it is a team issue and should have been handled within the team, and then talking to the press about it and expressing disappointment – how exactly is that a team move on the coach’s part?

Hopefully some coach will get on the right side of this issue.

CoreOS seems like a very nice idea – a minimal linux kernel with just enough libs to run and manage containers. So a container-ready linux kernel, systemd to launch and maintain containers, and etcd to communicate settings (key-value pairs) across containers. The project is motivated by high scale datacenter needs, but i think it is a much more general idea, i see applications in the appliance space as well – managing the software load on a widely deployed set of appliances is hard for many of the same reasons that managing the software load in a datacenter is hard.

I have been using a Centos kernel for my appliance experiments but am going to try a quick coreos experiment. There are some things I don’t understand yet – coreos at this point is focused on virtualized environments and the bare-metal support is just coming along. SystemD is cool but if the containers are centos based, not quite sure how that is going to work out. and networking within the appliance is confusing to me, the coreos docs suggest that one of the containers can manage all the traffic distribution but i am not quite clear on how that works at all levels – dhcp, dns, iptables, nginx, etc etc etc.

So lots to learn! Digging in…

OK so I brought up a CoreOS instance using Vagrant/virtualbox on my machine. I am trying to understand the network config and how traffic is managed and routed to the containers. From spelunking around the web, here is what I think is happening:

• CoreOS uses iptables to control routing and deliver to containers.
• By default the iptables rule set is totally open – everything accepted, everything permitted, everything routed. ie, /sbin/iptables -L indicates that there are no rules set and the defaults are ACCEPT.
• As an option, one could create an iptable firewall config and define a systemd config file that ran it. So system-wide restrictions can be implemented.
• Additionally/alternatively, the individual containers can define iptables firewalls (using systemd or whatever in those containers).
• you have a very permissive iptables setup in coreos, you are letting all
• OK besides the inbound ethernet interface, CoreOS defines a bridge interface for the containers to use.
• Per http://blog.docker.io/2013/03/docker-containers-can-haz-networking-now/, Docker analyses the bridge definition and uses that to assign IP addresses to the containers that start up. This is how containers get their IP addresses.
• The Docker command line can also be used to map certain inbound ports to the container. Docker will twiddle the nat table rules to achieve this.

OK so this makes sense. so now let’s look at what is actually happening on my system when I bring up CoreOS in a VM, with two containers, an ubuntu and a centos container.

• “/sbin/iptables -L” confirms that iptables is totally permissive. Check
• “/sbin/iptables -L -t nat” shows that Docker has indeed inserted itself into the nat table, the appropriate nat rules are there along with the necessary masquerade rule. Check.
• “/sbin/brctl show” lists two virtual ethernet connectors, one per running container. When I kill a container, one disappears. Check.

• “/bin/ifconfig” starts to confuse me a little. after the one container is killed, with only one running, i get the following interfaces:

  • bond0: I understand what a bonded interface is, but this seems like a rump inteface. No ethernet address, no aliasing i can find anywhere, there is nothing to bond? there is no packet traffic over the interface.
  • docker0: ok this makes sense, this is the previously defined bridge for the containers, check.
  • enp0s3: this has a reasonable looking IP address and is probably the virtual interface for the coreos host, check.
  • gre0: ok this is usually for ip tunneling, as is the next. But i am not tunneling anything I don’t think? And again no traffic on it.
  • tunl0. no traffic on it.
  • lo: standard loopback, check.
  • vethXXXX: this is the adapter for the active container, check.
  • within a container, ifconfig shows me the virtual ethernet interface and a loopback interface, and iptables is wide open. check.

So I think I understand the lay of the land. Maybe the tunneling and bonding are just created for all configs but only used in certain uses?

UPDATE: the good people on the coreos google group helped me out, the bond/gre/tunl adapters are temporary artifacts that are unneeded and will go away.

I’ve been an iPhone user since the first release, and have upgraded my hardware faithfully at every opportunity. I did take a little detour to Windows Phone land for a short while.

But I flipped this week to an LG G2 android device. I’ve been running the developer previews of iOS7 for months so there was no real new excitement from Apple for me this week on the software front. And the new iPhone hardware is fine but nothing stunning about it. And most importantly, neither the hardware nor the software addresses my number one smartphone problem: battery life. I have struggled with my iPhone 5, needing to keep charging setups at home, in all cars, at work, and cables and car chargers with me at all times. I use the internet a lot and I would get at most a half day of use on a charge. And then iOS7 just make it horrible with the background tasks – Facebook and Maps in particular would just kill my battery, and this never got a lot better during the iOS7 betas. I had to constantly monitor what tasks were running and manually kill them. Smarter battery management is sorely lacking on these devices – there is no reason that Facebook should be nattering away in the background burning battery when the phone is in my pocket. Maybe the app developer thinks their app is so awesome that it should do this, but the OS should be smarter about managing the tradeoff between network usage and battery life.

So I am switching. I am looking for the full smartphone experience with better battery life. Several Android phones promise this, and I decided to try the LG based on other’s recommendations. I’m only in day 2 so I can’t be sure if it has solved the problem, but it does seem at first impression to survive longer on a charge. And it certainly has every app I need. The level of polish is so much better than the Android devices I toyed with 2 years ago. Way too many buttons and options but I am learning my way around all that. The input keyboard is way better than iOS thanks to Swype and also better autocorrect preview.

I am diving into my new Android phone and there is a lot to like about it. I am not prepared to say it is huge winner on the battery life front, but I have hope there. But man the device really demonstrates the problems with the OEM model for product delivery.

• There is a ton of crapware from LG and ATT on the phone. that you have to shove off into a corner. ATT has like 12 apps alone, including their own app reco engine which is super sluggish. and either ATT or LG put something on here called “browser” and made it the default internet browser.
• Either LG or ATT decided to remove the Google search box from the homepage. and replace it with a weather widget. Awesome, thanks guys.
• There are like 9 places to put app shortcuts. It might be on your home screen. Or the notification bar that drops from the top that has some apps, and you can put more there. There are some app buttons jammed onto the lock screen that you can configure. ATT slams some sort of browser bar into some apps (I think this is ATT). there is some overlay of apps that comes up when you drag from the bottom of the display – is this android or lg or att or ? This is giving the Microsoft tile/menu/ribbon/charm morass a run for the money.
• Obviously google apps has to be in one of these places right? No. It is not anywhere on the phone. There is something called Polaris Office that ATT or LG jammed on that I don’t want.
• oh that is right, google renamed it google drive, so that must be on the phone! Uh, no. It is not. It is in the app store tho, and I can download it, and it seems to work fine. Why the f^&k isn’t it on the phone in the first place??
• @bradsilverberg mentioned that his new android phone doesn’t have visual voicemail and that verizon wants $3 a month to turn it on. I looked around my phone to figure out where voicemail is. Hmm, no voicemail button, nothing in the dialer about voicemail. So I call myself from another line and leave a voicemail. Notification arrives saying I have voicemail and that I have to dial some number to get it, and I do so and I am back in 1985 interacting with voicemail menus. Wow, is this really the experience??? So a quick web search and I find an “AT&T Visual Voicemail for Android” app that is free, from AT&T. I install it and it works fine. WHY THE F%^K IS THIS NOT INSTALLED BY DEFAULT???? Especially when AT&T slams on 10 other apps by default including their own browser, their own messaging app, their own mapping, etc…

I mean the parts are all there for a great out-of-box experience but as @bsilverberg said to me, someone needs to ride herd on this thing and clean all this crap up. It is just stupid to ship the phone this way.

Looking up at Attachmate world headquarters this morning on my way to coffee with an ex-MSFTer.

A good reminder that a) if you focus on a niche and keep hammering away, you can have a long long life, b) sometimes the boring unsexy niches are best, c) franchises can last a long long time if you tend to them.

• Reality Hunger by David Shields. An excellent exploration of what we read, why we read, the evolving nature of the things we read and the things we write. I used to think I was a thoughtful reader, now I realize I am a rank amateur. If you read extensively, I recommend this highly – you will gain insight into what you read, and into your own motivations for reading.
• Neptune’s Brood by Charles Stross. A lot of fun. Humanity merged with electronics in a fun way – adapted for life in space, water worlds, and other extreme environments. Also a nice exploration of the cultural and economic implications of a civilization spread across the stars, but having no FTL travel. In light of my first book, I wonder why I find hard SF engaging. On a positive note, I enjoy the complex thought exercises that a good author goes through to explore the implications of a certain technology or trend – it is just good brain exercise, keeps the brain nimble, encourages you to think longterm and to question preconceptions. On the negative side, what am I trying to escape from exactly?
• Medusa by Michael Dibdin. Very good detective tale set in Northern Italy, how did I not find Dibdin until now. Secret love and betrayal all hidden behind political intrigue. Again, tho, what am I trying to escape from? After a day of reading technical material and screwing around with open source software, these kinds of books tend to help me decompress, I think.
• The Rope by Nevada Barr. I am not very squeamish but this book is disturbing. I am unable to finish it, the kidnapping and brutality against the main character are just too disturbing. I’ve never had a problem with horror novels because the events always have an unreal, cartoonish feel. This book seems too real. Not my cup of tea.
• The Office of Mercy by Ariel Djanikian. Do not reward this author/publisher who are trying to ride on the coattails of the Hunger Games. A horrible book. My dog has more depth than the characters in this book. No, make that my dog’s chew toy. Really terrible.
• A Tap on the Window by Linwood Barclay. Starts out as a typical small town murder mystery: a mysterious encounter on a dark night, a murder, a long hidden family secret, corruption in the local police force and government, a PI with personal issues who works slightly outside the lines. I’ve read all this before. But the ending is not all neat and tidy, but is instead dark and emotional and painful. Nicely done. I’m not going to explore why I find a book with a sad, emotional, dark end so appealing.

Kickstart and Anaconda. Grub. Syslinux/isolinux. UEFI boot vs BIOS boot. USB key boot vs ISO image boot. Secure boot. Virtual Machine boot vs physicalized boot. Building packages. Integration with Puppet/Chef/Ansible/Salt. What a mess of configuration parameters, configuration files and syntaxes, constrained execution environments, etc. Design piled on top of design on top of design. Clearly the entire hairball should be reconsidered.

The most annoying impediment to deal with was the btusb bluetooth driver that singlehandedly would prevent the entire system from booting. Struggling to understand the design decisions that led to a system that can be taken down by a stupid race condition with the bluetooth driver. Bluetooth, for gosh sake.

This was a first for us. 15 minutes out of Anacortes and the captain announces we are taking a detour to help a vessel in distress. We change course and the search lights go out, and in about ten minutes we come across a small sailboat with sails ripped. It was a pretty windy night, probably too much weather for this boat. The ferry crew verified that the small boat was not taking on water, and so we just tracked him with our light for 15 minutes, awaiting the arrival of the Coast Guard 35 minutes away in Bellingham. Had the sailboat started to take on water, the ferry crew was ready to do an emergency rescue, but that would have been at some risk to the ferry crew and a lot of risk to the small sailboat, it certainly seems like the ferry would have crushed the boat.

Eventually another ferry arrived, and also kept their lights on the boat. The ferry captains decided for whatever reason that the 2nd ferry was the better one to track the boat, and so we moved on. The well-lit ferry boats had to be a welcome site to the small sailboat, it was a dark and windy night, had to be a little scary out there.

Read for instance Tales of an ex-Microsoft Manager or The Poisonous Employee-Ranking System That Helps Explain Microsoft’s Demise. (Aside: did someone at Slate get a shitty review years ago and now they are determined to get their payback?). If you believe these, Microsoft failed in part because of its corrosive review system, everybody hated it, not a single person at Microsoft liked it, its existence was hidden, etc etc etc. Man that place must have been utter hell to work at.

I was at Microsoft from 1988 to 1999 and have a different experience. Stack ranking (sometimes called the lifeboat drill) was in effect and the system worked fine. It wasn’t hidden, everyone knew they were being ranked. Yes there was stress around review season and yes people could be unhappy, but it didn’t immobilize the organizations I worked in, and people still worked hard and were committed. Perhaps the system in place at Microsoft in the last dozen years is dramatically different than the system in place when I was there.

In my working and academic life, stack ranking has been pervasive and omnipresent. I was stack ranked in high school via grades and class standing. I was stack ranked in college, every course I took had a curve. I was stack ranked in graduate school. I was stack ranked aggressively at Booz-Allen & Hamilton: it was an “up or out” organization, you either got promoted every two years or you were “counseled out”, and everyone knew exactly what everyone else was getting paid. I was stack ranked at Microsoft. We’ve kept track of individual partner returns at Ignition so I am effectively stack ranked now. The notion of stack ranking doesn’t offend me. The last time I wasn’t stack ranked was when I got my 5th grade “Certificate of Participation” for pony football.

Every team I have been on has a distribution of people with a distribution of effectiveness. The people at the bottom of the stack rank are not bad people, they are just the least effective people on the team. It could be a bad fit for them. Or maybe they had other things going on in their life this year that hampered their effectiveness. Whatever the reason, there are real differences across a team in performance and effectiveness. And some part (not all) of compensation will be tied to effectiveness, that seems pretty fair to me. Every manager and company I have worked for has been honest about this system and has explained it well.

My annual review results have never been a big surprise to me. And for a lot of people on my teams, the annual reviews were never a surprise to them either. Because annual results are not something that get unveiled once a year. You build your own annual results (and thereby your review outcomes) one month, one week, one day, one hour at a time. Your story is getting built all year long, and you should know where you stand all along the way. I was always very observant of my own results and those of my peers and tried to adapt along the way each year. People that aggressively manage their career and their work are generally not surprised by year end results.

So – for me anyway, stack ranking has been omnipresent, it is not irrational, it has never been an ugly surprise. But a lot of people think it is awful at Microsoft, and that eliminating it would have helped the company. I will offer a different view.

When things are going great at a company, as they were at Microsoft in the 90s, no one complains about stack ranking. When things aren’t going great, then people complain about stack ranking and about everything else. Why aren’t things going great?

• The mission for the company is unclear. Employees are not excited, not inspired. Customers are not excited and inspired. Developer interest is waning. Analysts are not excited, the company is no longer viewed as a bellweather.
• And growth in stock price has stalled, partly as a result of the lack of compelling mission. The rising tide is not lifting all boats. Comp starts to feel like a zero-sum game.
• And the organization is not growing dramatically. The opportunity to move around the company and find better fits is more limited. A low performer on one team may be a star elsewhere due to a better skill match, better team meshing, or whatever. Having a ton of new teams starting up is a great release valve to have in a company

If you took a magic wand and “fixed” the stack ranking system at Microsoft, but left everything else constant, people still wouldn’t be happy. The underlying problems of mission and growth need to be attacked, and the other dominos will fall. I suspect, as Ben Slivka has articulated, these problems are easier to solve in Microsoft broken into some parts with separable and clear missions. But the stack rank is not at the heart of Microsoft’s problems, and is not at the heart of issues to address going forward.