Guerilla Informatics

Difficult Conversations

I just got done listening to the best self-help book ever: Difficult Conversations: How to Discuss what Matters Most".

I highly recommend this book to everyone because it applies to both ones personal and professional life. From your neighbor whose dog barks all the time to the boss who you need to ask for a raise, this book helps you understand why such conversations are difficult and how you can make them easier - so that you don't avoid them! The books comes out of the work of the Harvard Negotiation Project.

Why are converstations difficult? Fear for consequences! Will I succeed or fail? Will I make it worse or better? What if I do nothing? "If its difficult its about you" - there's some personal reason that's making it difficult. The authors suggest that you should think of such situations as an opportunity. An opportunity to improve situations or strengthen relationships.

Here are a smattering of key phrases and concepts from the book:

• Questions disguised as statements: "Do you have to drive so fast?" really means "You make me nervous when you drive this fast". Which one is more likely to lead to a productive conversation?
• Multiple Stories: taking the perspective that you have your view or "story", the other person has theirs and there may be a third story that lies hidden under the surface. Exploring these stories is how you come to resolution off the conflict. Importantly, sometimes people just want to tell their story, and when they do, the conflict is resolved without further action.
• Absolute either-or thinking is a trap. Give or take is essential.
• Both stories matter. Both perspectives are correct. Do not fall into the trap of creating a discussion that is a zero sum game in which one party has to lose so that another can gain.
• "People don't change until they first feel understood"
• "Trying to change someone guarantees that no change will occur"
• Changing the goal of a conversation to be that of learning rather than debate of known points. "You don't know what you don't know" applies because their are usually hidden reasons for someones opposition to your position
• "Understanding doesn't mean you have to agree". You've still made ground if you understand each other, even if you don't agree. The goal of agreement as resolution can be a poor goal in many situations.
• "Identity Conversation": considering the impact on ones identity or self-esteem
• "Feeling Conversation": sharing feelings...one of the very few indisputable things that can lead to understanding
• "Learning Conversation": seeking and volunteering information
• It is important to re-position disagreement from a blame perspective to a contribution perspective. Usually both parties contribute to the problem...sometimes by avoidance of it...sometimes by subversion...owning up to your contribution is important.
• Watch out for statements that can cause an "identity quake": a message that can be received as a criticism that shakes someone's self-esteem. For example in parenting, anything that is critical of certain important parenting choices can be received in a very negative way. To accept the statement as true implies that you are not a good parent, which may be something profoundly important to your indentity

mac development juice

Just back from WWDC with some comments on OSX development tools.

Making great applications (small and big) for the mac is a lot easier now with Tiger development tools. Sometimes its the innovative aspects of of these tools that matters most. Other times its just the fact that they are bundled into the operating system that is important because it shows Apples commitment to improving them.

With the Intel announcement, Universal Binaries are essential, and I expect this to work as advertised. Distribution for these new binaries is troublesome, however, and I wish Apple would open up there "Software Update" system so that other vendors could use it to notify the user of updates. Some vendors will exploit this to get upgrade fees from their customers, however, so Apple should do what they can to discourage this practice.

Code translation with Rosetta becomes important during the Intel transition, but if it works OK, then I think most folks will tolerate it. The important change here is that Classic apps will no longer be supported

Brent Simmon's called Automator the sleeper hit of Tiger, and I have to agree with this. It breaks down the boundaries of applications and provides a cooperative environment for performing a wide variety of computational workflows. It is important because it allows people to automate tasks they already do manually - this alone is significant. To the extent that a workgroup can model their workflow with actions, Automator lets them create applications that run it, enforce policies, and notify users about progress. The key is that they won't need a programmer to do most of what they want - just one tech-saavy mac user in the group can program the workflow and since its self-explanatory, others could alter/improve it over time. This will be a huge hit in scientific areas where custom workflows and very common. I hope to see branching in Automator 1.1, however, so that workflows can be more complex and not require use of non-workflow programming to handle exceptional data.

Core Data schema and form designs tools are very nice. ER modeling and form development tools are pretty much what you'd expect from these. Tight integration with XCode and not having to use Oracle are big wins.

For Java developers, the Hotspot C2 Compiler for Intel means a bump up in speed for longer running programs (2.3X on Scimark 2.0 benchmark), and other intel-based optimizations to come.

Dashboard widgets are for "Marketing and Promotion" says one friend (I agree). Its a way for companies to advertise their internet services on the Mac, without having to learn Cocoa. It will also provide seed applications for Apple's future foray into home info-entertainment products.

post wwdc predictions for apple

Intel boxes targeted as home entertainment hub. Mac mini-sized with innovative display, remote control, and hardware-based digital rights management ( DRM mention in this NYTimes article)

Multi-user Spotlight search of shared drives.

Branching in automater. Possible support of workflows with cyclical graphs.

Breathtakingly innovative and beautiful Intel-based powerbook (Xmas 2005 or shortly after).

Enterprise compatability hooks: tighter MS active directory integration (through LDAP), Lotus notes itegration?

Web service libraries that supoort richer and more powerful Dashboard widgets and Automater actions.

Dashboard widgets targetted for home appliance product, possibly a small screen device.

Ability to run older windows programs in Mac OSX w/o purchasing windows. Wine (sun partnership?) or other technology with cleanroom win32 API.

WWDC: The Message about Intel, Universal Binaries, and various platitudes

I spent the last two days at the Apple WWDC getting The Message about Intel, Universal Binaries, and various platitudes about the transition.

On the mac developer side of things, the Apple engineers seem to be easing their concern by helping them recompile their code on-site during the conference..."see, now that wasn't so bad was, it". My Apple friend says that he's seen very few problems, but there's always a few stories in the air about the poor sucker that optimized his code for PPC. One response I got was simply that the PowerPC is still a great platform and that these investments will continue to pay off during the transition period (2 years?).

Suppose that 95% of developers try out the new tools, decide that the technical hurdles are minor. What about their customers? What about increasing market share? Uncertainty is never good for purchase decisions, so surely it will hurt, but does the chip really matter that much any more? For desktop PCs and laptops, this move is really about making better computers that run faster and cooler. Apple has delivered this in the past and customers expect them to continue to do this. If the method is Intel, then fine. Meanwhile, many of us will benefit from cheaper hardware prices caused by concern that bears little impact on use of a computer under a 3 year life-cycle. Universal binaries should be ubiquituous in two years, so all the applications we use (and their upgrades) will run fine on such computers.

The HPC purchasers, however, may really jump ship. Buying an XServe grid at this point seems like a bad idea. Its probably best to wait a year for the Intel boxes and recompiled apps. These customers are far more sensitive to changes in performance and have a somewhat longer planning horizon. And why would they want further heterogeneity of both PPC and Intel boxes in their grid. Another thing to worry about. Watch prices drop even more dramatically here.

One good thing that I'd like to end with is that this will be a boon for Java developers on the mac. Apple reports that the C2 Intel JVM is 2X+ faster on some benchmarks and their JVMs are sure to be better and faster on Intel because of the broader user base.

Footpaths and schemas

Footpaths and schemas: "

I'm sure there are dozens of versions of this story, but I heard it from Larry Wall, the father of Perl, and it goes like this. Instead of laying down sidewalks, the builders of a new university campus waited for footpaths to emerge on the lawns. Then they paved the footpaths. Perl's design was informed by this idea of structure emerging from use, but that was an unusual case. We typically lay down the sidewalks first, and when footpaths emerge we profess surprise or try to ignore them.
...
Of course it's easy to criticize information systems that fail to embrace sloppiness. It's much harder to explain how they should. Sloppiness is only a means to an end. In order to make things work and get things done, we need to codify patterns of use. It's a Catch-22, though. The right patterns don't emerge from systems that people won't use. How we reconcile specification with emergence isn't an engineering discipline, but it probably should be. [Full story at InfoWorld.com]

..."

(Via Jon's Radio.)

broadband matters

I'm a full-time telecommuter and have recently signed up for Verizon's fiber optic line to my home which gives me about 5x faster speed downloads than cable and about 10x faster uploads (for the same price, but even faster speeds are higher rates are available).

The guests on NPR's On Point a few weeks ago talked about the impact broadband has on telecommuting (or other remote work), and I believe this is the key area that we are enabling with this infrastructure.

The providers of broadband service, however, are focused largely on providing big pipes to sell TV programs and movies -- their focus is on entertainment, not work.

My new fiber optic service will probably result it me canceling NetFlix (DVDs by mail) and instead getting direct movie downloads (and more of them). Multiply this across the board and its a big win for Telcos and Hollywood. But is this a significant benefit to society?

What's really interesting to me is the 10x increase in UPLOAD speed. What am I going to do with that? I really don't know, yet, but there are a variety of peer-to-peer services (e.g. BitTorrent) showing up where I can participate in the distribution of content (legally). I can also easily run several phone VOIP lines for my business, and use video chat, but I'm not even touching the potential here. The important point is that it allows me to work better and _participate_ more actively in the global network.

My hope is that widespread adoption of broadband will have a far greater impact on the way we work than the way we are entertained. If people can work better at home at jobs with more independence and flexibility, then this can really improve peoples lives. It may allow some familes to break out of the two income trap, and work 1.5 jobs total for two people instead of 2.5 with no time left for kids and enjoyment of life.

Ubiquity of these types of connections are also very important, and I think the government has an important role to play here. The internet is a human communication medium, and the human-ness increases with greater bandwidth and full participation. Video chatting and more immersive technologies have the potential to allow us to feel closer to others and thats good for all human endeavors. Reaching anybody and having everyone part of the conversation is essential.

Stickybrain update and .mac offer

I've been evaluating StickyBrain over the past month, and it just expired on me. I was reluctant to purchase because it costs $40 and the previous version required the Missing Synch $40 for palm synching, but this .mac discount puts it at $20, and the new version has its own Palm conduit. Apparently its got a new database for Tiger, but I don't know what the benefits of that are.

Anybody who has .mac and a palm device and wants synchronized list management software should get this. It also has .mac backup and iPod synch if that matters to you.

The rise of grassroots media networks

Got free or cheap video content? You should have little trouble finding someone who'd like to host it for you. Try BrightCove, Google, or just announced from Andreesen/Homer: Open Media Networks.

Uber Pipe and PDA

This past week, I've made some big changes in my digital life. This post can serve as an addendum to my previous survey of Free and cheap small business telecommunications.

I finally got the Treo 650 that I've coveted for over a year. I'm using Sprint because I live in a gully that verizon/cingular signals can't seem to reach. I got a used Treo on ebay that came with Kinoma producer/reader for putting movies onto this thing. Rather psyched to have this, but my kids might use it more than I for long car trips. As for the service, it turns out that if go for the 2000 minute plan, you get lots of stuff free...add a phone, unlimited internet. It was a small victory for me to walk out of the Sprint PCS store with tons more features and minutes and not have to pay any more per month.

The other big event was finding out that those Verizon trucks in my neighborhood over the past month have been laying fiber optic cable for a consumer internet service named Verizon FIOS. $39 for 5MB down, 2MB up and $49 for 15MB down and $15MB down. Compare this with (my current service) Comcast at $45 for 3mb/300K or $55 for 4mb/600k (or something close to that). Yeah, you bet I signed up! Now all I need is for Apple to come out with their movie download service.

Intermediation, Attention, and Social Software

Jon points out in his architecture of intermediation piece that the web needs a general purpose way for people to separate public and private assertions such as de.licio.us tags.

Another type of public/private assertion is simply what you've been reading or more generally what you've been paying attention to .

With such tools in place, individuals could be producing a stream of knowledge about their activities as they go about their daily business. Such streams could secured, filtered, edited, and broadcast in a semi-automated way.

It all comes down to lowering the activation energy for sharing your information trail with others in a way that you are both comfortable with and benefit from.