ESVBibleWorkflow – This is an Alfred 2 workflow to place scripture from either the ESV or NASB on your clipboard. Super useful (for me, anyway). I like that if I can remember part of a verse, but not which verse it is, I can start typing and it will search ahead for me. 1
Slogger – I use Day One to journal a variety of things which make no earthly sense to share on this site.2 Slogger grabs my content from various sources3 and creates Day One journal entries for me. When I can’t journal or forget to journal, this does it for me. It’s not a good excuse, but it grabs content I’m already publishing elsewhere. Like this post.
Technical Difficulties – Like a normal podcast, but broken… these guys talk for about an hour about a relevant, important, nerdy topic, and bring in great guests to discuss them when needed. The recent episiode with Bradley Chambers about Wi-Fi was incredible. I’ll likely listen a second time. Also, their show notes reset the standard for podcast show notes.4
Because for some reason I can never remember that “Woe to you, when all people speak well of you, for so their fathers did to the false prophets.” is Luke 6:26, but can always remember “speak well of you” is part of it, so this is useful. ↩
There’s a strong argument going on in my head to drop this blog altogether. The nerd in me that started writing online in 1999 can’t do that just yet. ↩
All sorts of social sites are supported. For me, I pull content from here, instagram, pinboard, pocket, last.fm and twitter. ↩
I used to be such a good nerd.
Then I had two kids.
Somehow, between a wife and two boys, all of the time I used to spend pouring myself into web design, picking up a new programming language, waxing poetic about the latest nerd-rage-inducing article, or tinkering with some piece of hardware, has been replaced with other, far more valuable things like fixing my house, taking a son out for a donut, going apple picking, reading a bedtime story, bouncing the baby, etc.
There's no question whatsoever that my priorities are right in that regard, but when the technological infrastructure I set up when we first bought the house starts to break down, I have to fall back on admittedly somewhat atrophied muscles to fix them. This was the case last month when the state of our home network drove my wife to declare that I should not send her email during the day because she coulnd't read it anyway.
I had, of course, noticed some problems with our home computer as well. But, in my defense, I rarely sat at it, and when I work from home, my MacBook Pro has had no problems with connectivity. Well, the wired network was flaky, but wireless was fine, so I guess… I guess I was just putting off figuring it out.
My home network is quite simple. I have a Verizon FIOS (Actiontec v1) router which has a 4-port 10/100 switch and an 802.11b/g access point. The access point is disabled, and one of the four ports has a Linksys WRT610n 802.11n router with 4-port 10/100/1000 switch connected to it. The Linksys has DHCP disabled and a static IP assigned outside of the Actiontec's DHCP range. It's an access point and switch. It serves up both 2.4Ghz and 5Ghz 802.11n SSIDs using WPA2. Good enough, and it's been working for 4ish years.2
There are three machines connected to the Linksys' wired switch:
All other devices (iPhones, iPads, MBP when anywhere else, wireless printers, guest computers, etc.) use the 802.11n network. The signal strength is sufficient in the house, but anywhere outside gets nothing due to a nice, solid concrete/brick foundation and walls.
About 3 months ago, I noticed that the mini was taking a long time to render web pages. I figured it was a browser thing, because it was sporadic. I cleaned the machine to the best of my ability, trimmed down browser plugins, made my wife stop leaving 8 tabs open, etc. It seemed a little better, but something was still off. I switched it over to the wireless network, and things improved dramatically. I chalked it up to wiring or something and went back about my other family business.3
Then it came back, while on wireless, but again, sporadic. I started using speedtest.net via speedtest-cli and noticed that I was getting about 3Mb down and 1Mb up4 on both the wired and wireless networks. From my MBP, I got 50Mb down and 30Mb up when on wireless. It had to be the mini. I was running OSX 10.7.5, had never done a wipe of the drive since upgrading it from 10.6, and who knows what sorts of cruft had been building up over the 3 years since we bought it.
Then I started seeing the same thing on the wired network for my MBP, which was fishy.
Perhaps I had two problems? Our wiring was bad / the switch was bad, and the mini was just in bad shape.
Rather than back everything up, format and reinstall, I bought a fresh 500Gb 7200rpm drive, installed Mavericks from my MBP and swapped the drives in the mini.5 Suddenly, it was back to normal, even wired. Crisis averted.6 I mentioned to Cait that she should watch some youtube videos to see for herself, but she quickly forgot.
Two days later it all came back.7
Not knowing why this was all happening had finally started bothering me enough to feel the need to cheat on my evening time with my wife to improve things. I started by isolating the problem:
Based on all of this, the only conclusion I had was that the Actiontec router itself was having trouble. That, or I was somehow being throttled based on MAC address only, but that sounded fishy. And supposedly Verizon doesn't do that, so yeah, router.
I have never felt like Verizon was all that great with customer service. In terms of cellular service, I've never really had the need to interact with them, and when I did I often felt like they didn't much care about me or my issues, just getting me out of their sight. I figured that if I called FIOS support, they'd need me to repeat everything I did already, but I had no choice left.
Thankfully, I got on the phone with someone who spoke my language. I re-iterated the problem, as well as what I had done to troubleshoot it for myself. The tech asked me if I had tried a few other things which I had forgotten to mention (I had), and then said “Well, you have a V1 router. It could just be broken. I'll send you a new one next day air.”
He walked me through a manual release of the current router's IP address and said to skip the install CD (it's Windows-only anyway) and just wait 5 minutes after the IP release to turn on the new one. The router was there the next day, I did all that he suggested and we're back to our normal blazing speeds. The new router has 802.11n and a 10/100/1000 switch, eliminating the need to use the Linksys. It's coverage is incredible. I was going to root the Linksys and turn it into a repeater, but there's no need to do so. I can use wi-fi in my garage now, which had been impossible before.
The title of this post is a hat tip to Ian Beyer who used to have a blog by the same name. I'm sure if he and I lived closer to one another he would have had this figured out in minutes.
We used to use the Actiontec's wireless network, but when Circuit City was going out of business a few years ago, I picked up the Linksys router during the firesale so that we could get a speed boost. The gigabit switch was worth having almost more than the 802.11n coverage. It's running the latest firmware that
Linksys Cisco makes because, at the time, DD-WRT and OpenWRT did not support it, though they both do now. ↩
See, this is what I'm talking about. I never would have let this slide back in the day. ↩
We have the FIOS Quantum 50/25 package, so this was notably unacceptable. Again, 3/1 is pretty fast for the days of my youth, but it was relatively high-latency as well, so it wasn't like things were only slow, they were also wicked delayed. ↩
Never, ever think that anything is so simple. Ever. ↩
Cait was pretty annoyed that she never opened youtube in those two days. ↩
It's only wired, but it at least showed I'd get the same results with both OSs. ↩
I rarely leave home without any of these things. I took a few months off from carring the multitool this year right after Thing #2 was born this summer, and at least once a week I’d wish I had it on me.
I think I’ve reached the bottom of my EDC minimization. I’d love to reduce my keychain a bit, and may try removing the smaller keys (bike lock, for instance) and the key to my work office that I never lock. That doesn’t seem like much, but every ounce counts if it’s clipped to your belt.
I've been following The Sweet Setup since it started a few weeks ago. For the most part, I've 100% agreed with their App preferences, and I think my digital workflow is fairly well-streamlined now anyway. 1 I've recently added a keyboard case to my iPad, which has turned it into an incredibly convenient bedside-table computer and journaling/blogging interface.
I largely don't blog any more, partly due to a lack of time and partly due to an inner struggle around whether or not any of this content really matters; questions like “Is this site more than just simple vanity?”. I'll presume it's not for a little while longer and see what comes of it.
One of the apps recently reviewed by a Sweet Setup post was Editorial, which is a full-features text and Markdown editor for iPad. It's pricey ($10), but the reviews and demos were very compelling so I thought I'd commit some time to playing with it.
Goodness. It's incredible. By far it's the best-looking and most easily usable Markdown app I've tried to date. I'm only touching on the workflow support by using it to write this blog post, but I can see myself doing much more with it in the future.
Boy am I tired.
The day got off to a really good start. I made my rounds through the vendor booths again, and spent five minutes with each of the testing vendors asking them a few specific questions:
I gathered some good data and will follow-up with a few after talking to our QA Manager when I get back to work.
I spoke with a few vendors I couldn’t get to yesterday and ended up two more shirts heavier including a sweet hoodie from the fine folks at RefinedWiki. (Thanks!)
My plan had been to hit several presentations, and that sort of worked…
This was actually a really fantastic presentation that gave me some good ideas to kick around at the office. I’m seeing some of my newer engineers start to want to flex their muscles a little more, and ideas like a quarterly hacking day, or some way to promote innovation time outside of a given sprint lead me to think this might be a smart use of our time. One other stellar suggestion was to do brown bags, which we’ve done before, but when you start to burn out, use the time to watch presentations from relevant conferences. This was one of those DUH! moments — I can see myself showing at least half a dozen from this conference to my dev ops folks and project managers.
I was not expecting much from this session, but I was also pretty pleased.
This was identical to the webinar I watched on the plane flight to San Francisco, which you can find here. I tuned out and did some release documentation. Good use of time, actually.
This session was so full I couldn’t find a chair and had to leave. I tried to extend this into a cat nap, but couldn’t turn my brain off. Instead I found a table, put City on a Hill on repeat, and banged out some more documentation.
I’ll have to get the recording and watch it later. There are several others I want to see as well.
Presentation was hard to follow, and actually I got bored really quickly. No biggie. Folks were getting tired. Instead I worked on some new JIRA filters and daily report subscriptions, and played around with some of my mac defaults.
And updated this blog post
This was awesome. Seriously awesome. Six engineers presented their most recent Ship-It! projects, and we got to vote on which was the best. They were:
The process was very cool. I’m interested to see what pieces of these things are portable to my org.
Yesterday I spent all day talking about Atlassian software, where my team was struggling, where we were succeeding, and what we wanted to do next. For the most part, the ways in which we are struggling are small and easy to identify:
There is no Atlassian solution to #1. (Sorry, self.) I spoke with the Product Manager for Development Tools Integrations last night on the bus to the “Summit Bash”, and she had no suggestions. I don’t blame her for this, of course. I work for a software company also, and we can’t magically solve infrastructure problems either.2 She, and the Product Manager for Stash, both told me that there was a lot I could still do with linking between the Dev Tools (JIRA, Bamboo, Crucible, etc.) and our SVN repository. No awesome branching, but that’s an SVN problem, not an Atlassian problem.
I’m going to need to write up an entirely different post for #2. It’s a big challenge for us.
Other customers with whom I spoke had different suggestions for ramp-up training:
I’m not sure what we’ll do yet. Consulting-based training is expensive, though less so than I had imagined.4 I only employ two Operations Engineers, and this isn’t my full-time job (some days, though…), so we’ll have to see what the company is willing to spend money on.
I had intended to hit up four5 different sessions yesterday:
This presentation was given by Stefan Saasen, a Developer Advocate at Atlassian. The general topic was about how the teams there made the decision to switch from SVN to Git in 2011, and have had incredible benefits as a result. He showed us several good resources for research and several good resources on how to do the conversion itself. It was a nice primer, and answered questions I would have had to ask in subsequent sessions.
So, you know, good programming decision.
The panel was so full I got kicked out due to fire code concerns. So I went and did an on-camera interview for Atlassian about how we do Agile, what issues we’re having, etc. I got a(nother) free T-Shirt out of the deal, so you know that’s cool and all.
This presentation was really less helpful than I was expecting. While it’s a big concern for us, I already knew everything that was presented. Once we solve our internal issues, we’ll reap some great benefits.
Best presentation of the day. Alex Holtz, from Orbitz, went through their entire migration strategy and execution from a commercial VCS to Git over the course of 11 months. Great anectdotes, great war stories, etc. He showed us the why and the how and explained some of the pitfalls associated with git and how you could (easily) avoid them. They chose an open-source git server called Gitorious, and were just starting their migration when Stash was announced last year. They plan to move to Stash in the future.
Given by two of the Dev Tools Product Managers, it was more of a deep dive into things covered in the keynote about development process and the concept of taking the “work” out of “workflow”. Nicely, though, they also talked about being productive with SVN in the meantime, which I really appreciated since, well, my team has to be.
So today, my plan is to try to get to these:
Hopefully today is as productive as yesterday was. Hopefully there are no more T-Shirts. My bag is full. I’m going to have to start handing them out to strangers, which might be a good idea anyway.
I’m not really opposed to this, of course. And as I said yesterday, I have a burgeouning obsession with git. Really, I don’t know what is taking me so long, but I digress. Which is what footnotes are for. ANYWAY. ↩
And there are several present who are happy to try to sell you on their services. ↩
I spoke with the folks from cPrime for a few minutes who estimated that a day’s training on my topics for a team of any size would be roughly $2500. I spend that much to send one person to a 3-day course on one topic. ↩
Actually it was five, but somehow I missed the one from Orbitz when I posted yesterday. My bad, my bad. ↩
Hey, I manage a team of really smart people, but I’m busy and I know I could be doing more tomake them even better. Seemed relevant to my interests. ↩
I’m still operating on Eastern time and stayed up late to check into my flight last night. I likely will need a nap to prep for my redeye. ↩
The last time I went to a tech conference, it was 2005 in Atlanta, GA for the annual ResNet Symposium. I loved that group of people. Over three years of attending, I had developed a network of about a half-dozen academic technology peers who faced all of the same challenges I did in that job. We kept up throughout the year, supported each other’s presentations at subsequent conferences, and generally enjoyed each other’s company when we had the opportunity to connect.
I left the world of academic technology later that year, and as such never had the resaon to attend again, and in my following positions, there had not been any opportunity to meet up with peers in whichever role I had. As a technical support engineer and manager, the available conferences all felt stale and boring,1 so I never even tried. As an engineering manager who was never a software engieer, I’ve often felt a bit out of place a technical conferences. I’m super happy to send my engineers to No Fluff Just Stuff any chance I get, or send my UI lead to the Rich Web Experience. They always come back with a slew of ideas, and it’s well-worth company dollars to send them. Sending me, though, would not be quite as useful.
This year, though, that got to change. My department within Nuance Healthcare went through a massive re-organization several months ago which separated us from our ancestral department. As a result, we took the opportunity to look for ways to be more efficient in our day-to-day operations that were not necessarily tied to the tools and methodologies of the previous organization. That org had moved all engineering teams to Rational Team Concert about a year beforehand, and while some were finding it useful, my team found that it just got in their way any time that it could. So, I set out to replace this with something else. After (some) deliberation, we pitched moving to the Atlassian tool suite, namely JIRA and
GreenHopper Jira Agile.
The Atlassian Suite is really easy to swallow as an investment: It has a huge community, has great name recognition, works well, looks good, is constantly being improved, and has a price tag that makes most corporate entities chuckle and as what the catch is.2 There isn’t one. It’s a stellar platform that “just works” for nearly any engineering (and non-engineering) workflow. Heck, NASA uses it.3
We’re working towards rolling out most of the suite to our entire department, which is about 290 people, with an expanded group of “external” users, getting us up into the 400ish range on JIRA and Confluence, with our engineers on Crucible as well. We tie all of these tools together to our Active Directory using the Crowd tool for Single Sign-On. It’s smooth, usable, intuitive and functional. So incredibly funtional.
Anyway, back to the point. Atlassian hosts an annual user’s conference called the Atlassian Summit, to which I was incredibly fortunate to be allowed to attend this year. I flew in yesterday, registered and wandered around the vendor booths for a bit before the opening reception. Today is when the main event kicks off, but I already feel like I’ve been exposed to enough to have been worth the price of admission. What I learn today and tomorrow should be enough to convince the company to send more than just me back next year, or so I hope. I’m planning on attending the following sessions today:
More to follow, I imagine. As for now, I’m still operating on Eastern Daylight, so I’ve been awake for almost four hours already and it’s not even time for breakfast here. Sigh.
Realistically, any event that gets me more junk mail from HDI is something I’m going to avoid like the plague. ↩
For the record, I don’t actually think their software is over-priced. It’s perfectly positioned such that small teams can afford it, and large organizations can afford to make the investment without an incredible amount of deliberation. It’s a lot easier to budget for a few thousand dollars than tens of thousands of dollars. ↩
Quick aside: My first engineer hire was in December 2007, and he still works for me (which is saying something about him as nobody else that I hired for that team still is, but I digress…). Within days of starting, he was trying to sell me on JIRA, but the company had just finished migrating off of FogBugz to DevTrack (huge mistake, by the way), and so that went nowhere. When we looked to replace DevTrack, there was no discussion about alternatives to what another group at Nuance was using (Test Track Pro, arguably also another big mistake), but he kept bringing it up. When we were forced to use Rational Team Concert (an even BIGGER blunder), he brought it up even more. At one point it was a weekly mention in our team meetings. As soon as the split happened, he walked into my office and said “so can we buy JIRA now”. It was incredible for both of us for me to say “actually, probably”. I have not regretted a single day after using it. ↩
I’m becoming obsessed with Git. More on this at some point. ↩
It’s always nice to hear other people’s ideas on how a team can be and do Agile more effectively. Continuous Improvement is the name of the game. ↩
This is one area my team is actively trying to improve, so I’d like to see for what we could be aiming. ↩
The Design Axioms describe the minimal rule set for designing interfaces: the 16 foundational concepts that are required knowledge for engineers and designers to create usable and elegant interfaces. The design of the content (and book) will follow the same tenets outlined within it. Diagrams, screenshots, and pictures (the visual story) dominate the content; text-based descriptions are short and to the point…
Our User Experience lead got a set of these from a conference she recently attended, and I fell in love with the cards so much that she got me a set. This one is my favorite because in no uncertain terms, I hate Gantt Charts. They are terrible, horrible, no good and very bad.1