Rambling Ken

http://blogs.sun.com/wallich/date/20080322 Saturday March 22, 2008

SXSW Roundup


Well, this took a little while, but figured it was a good time to give a quick roundup of my SXSW trek. I also picked up the SXSW Plague (the flu that's making it's rounds all over the country), hit me as I left the conference, and kept me in bed for a couple days. Next year, antibacterial hand wash in my backpack!

SXSW is actually 3 different conferences (actually, almost 4), co-located, related in some ways, with several overlapping. There's Interactive (my main reason for going), Film, Music, and several sessions and a room dedicated to video games as well.

The first thing for a newbie is to not try and do everything, but rather go with the flow. It's OK to miss things, otherwise, you'll just go nuts, and not get much out of it. A challenge for overanalytical planners like me! Also, doesn't hurt to pre-plan. Check out Upcomming, go the SXSW site and make a calendar of things to do, when you get there, take the "party" envelopes and mark down what you want to do, then take it easy, and go with the flow. You're likely to meet some group of people, then switch your plans based on other impromptu parties. And, it almost goes without saying that if you're not into social networking, you won't have nearly as much fun. It's all about contacts, meeting, improvising.

During the day, there are a huge number of sessions, at any one time I'd find 1-4 sessions I wanted to be at. I find the best purpose of the sessions is to identify people you'd want to catch up with after, and chat. Although some of the sessions were excellent on their own, I found them most useful as a basis for discussion with people later.

In fact, that's the most important to remember. The conference's real value is in networking. In the hall, at lunch, at dinner, and most importantly at the series of parties into the night. The best conversations and meetups I had were generally sometime between 11p-2am! And, at SXSW, everyone is up for chatting. Just go up to any random person (or someone you've always wanted to meet), and say "hey". I met a ton of folks, many I'd so far met only on Twitter, or reading blogs or found via other social networking routes. Also, plan on staying up late, this isn't an option, you just won't get much of of the event if you treat it like an "old school" tech conference. The first sessions in the morning are a 10am, and they tend to be sparsely attended as people wander out of the hotels kinda late. Not being as young as I once was, I alternated late nights, getting to bed by midnight every other night, so plan ahead, knowing you'll be up late if you're getting the most value from your time there.

One challenge was, since film and interactive happen at the same time, and there are a lot of movies playing at night, what to do? I highly recommend if you go, and you go to film as well, save the nights for networking, do movies during the day or early evening. And, have fun! Again, you'll get out of the experience what you put in. You're not there to consume, but to participate!

Music is another experience altogether. I only had time for one day at the Music conference, but it felt like the number of people doubled or tripled, I think that's because Music takes place mostly in the clubs, and not in the convention center, so the streets immediately fill up. Next year, I think I'll take some vacation directly after so I can visit more of the music. 1500 bands in tons of locations, much like the sessions at interactive, it's just an experience to be immersed in, not to try to manage.


http://blogs.sun.com/wallich/date/20080227 Wednesday February 27, 2008

My new role


I recently took a new role at Sun. Previously I'd been looking at Software Appliances, and ways we could use that to package various Sun technologies, and ways to build pre-packaged, managable software entities. Fun work, great research, and since I started it, several projects also started implementing the building blocks to make that happen, such as all the work going into packaging of OpenSolaris, and the many Software Appliances becoming available in the OpenSolaris repository.

I'm starting a new challenge, which is leading the JavaFX efforts at Sun. This is an exciting new role for me, with ties to my prior tools background in Sun and other companies, as well as consumer/end-user facing products, including mobile and other devices, and the overall User Experience for Sun's software projects, areas I've both worked in over my career, and have a continuing passion for.

I've got a bunch of embrionic ideas, and am finding that people on my team also are a trove of great ideas, passion, and excitment around what we're building, and what we can build. It's energizing to feel the level of interest in all the areas we're working on. It's a huge challenge, logisticstics, roadmap, ecosystem all being worked on, which is what makes it so energizing.

So, look for more frequent updates here, no, really. I plan to post what we're doing, where we're going, and what areas I'm currently excited about.

Next week I'm going to SXSW, not just because I am personally interested, having wanted to go in prior years as well, but now as an interested participant for my day job in terms of social interaction and media and music and video. All areas that touch on where we'll be working on JavaFX, and especially working with others on how they can express themselves, and their creativity using a powerful, highly interactive, multi-platform, multi-screen solution.

Expect to see an exploration of the possibilities a JavaOne in May, and also updates from me here on what we're doing!

http://blogs.sun.com/wallich/date/20071005 Friday October 05, 2007

That iPhone update thing

Now that the furor has turned into a low growl, a few thoughts. I'd "hacked" my phone with iFuntastic. Mostly because I missed the ringtones on my E61, where I could pick any media file on my phone to play. So, I tossed in a few of those ringtones. Yeah, I get that the RIAA figured out there was big money in ringtones, so they came up with special licensing to extract money for that too. However, because of the RIAA's heavy handed tactics, and actively attacking customers, informing us that paying for music doesn't mean we actually own how we use it. So, most of my ringtones are from non-RIAA sources. I have several friends who are musical artists, I have their "personal label" CD's, and their express permission to use their music however I want, short of reselling it. And, some are from old recordings for which no ringtone rights have been negotiated. So, if the music isn't popular enough to license as a ringtone, does that means it's explicitly not licensed as a ringtone? I don't think so. I don't see any EULA in my CD packet here for my Mozart symphonies. But, Apple needs to placate the mesozoic media companies, so I guess it had to make deals with the devil to agree to disallow any random ringtone creation. Anywho, so I did that. Then folks figured less arduous ways of putting ringtones on the phone. That was cool, but I already had mine set up, so I didn't mess with that. Then the infamous 1.1.1 came out. Now I was a little nervous. I knew I'd eventually want to upgrade, but now I might have a pending brick. Now, I think it was a poor marketing choice for Apple to go this route, but I don't see anything inherently evil. They came out with an update, they knew it would and in some cases would brick phones, so they warned you before you updated. Threw in a special "Danger, Will Robinson!" dialog and everything. At that point you could a) say to yourself "self, I modified this phone, do I really want to risk updating", or b) take your chances. When I bought the phone, I got a device, and a set of software with it. That's all I got. Folks found a way to hack that version of the software, great, if I wanted to use it, that's my choice. I don't have any illusion that I bought that and the right to tweek it, and the right to free updates from Apple. If I want to use a device, with the restrictions that were made clear to me when I bought it, that's my choice, and if I choose to disregard those restrictions, which I often do, well, you break it, you lost it. The train of thought that you bought the phone, so not only can you hack it, but that Apple owes you compatibility, and owes you not to unhack it? Sony has been battling back and forth with the PSP for a very long time, and DirecTV had a similar battle with people unlocking all the channels for a long time, until that got pretty difficult to pull off as well. Not everything you buy is a general purpose computing device. A Palm, mostly a general purpose device, other smartphones? Kinda similar. The iPhone, well, not so much. And, if it wasn't so absolutely brilliant at doing what it does do, then we wouldn't really care, now would we? So, how about this, if you want to hack it, fine, hack away, but don't update until the hackers catch up with Apple, and if Apple does something that makes future updates just not work, well, enjoy what you've got. Works for the PSP folks, should work for the iPhone too. Now, does this cause Apple some headaches? Well, sure, when there's some real hardware/software competition to the iPhone, and it's open and developer/hacker friendly, then there will be some migration. But, remember, Apple didn't build and market the phone to developers/hackers. Just like the iPod wasn't opened up, and yet became the largest selling music player, they're betting that same thing will work with the iPhone. And, for non-power users, that want power a different way, they may be right. I'm gonna enjoy the heck out of it, just like I did my E61, until the NEXT killer device/platform comes along! I'm just excited that Apple kicked the mobile industry in the butt with a platform that could be built today, if only other providers had thought out of the box enough to do it. The iPhone is, as a friend told me, "The Mac on a phone", it's kinda like Frontpage. Apple's got a way they want to interact with Consumers, and a lot of it is in limiting the customization. That's got a similarity to my day job, looking at ways to make general purpose hardware and OS's more like an "Appliance", once you know what you want the device to do.

Powered by ScribeFire.

http://blogs.sun.com/wallich/date/20070719 Thursday July 19, 2007

Flock 0.9 posting test

Testing with Flock 0.9, making sure the account setups worked...

Blogged with Flock

http://blogs.sun.com/wallich/date/20070625 Monday June 25, 2007

Upgrading and Software Appliance goodness


For my personal web hosting, I use a service provider (Bluehost), and in the spirit of using a hosting provider, try to use their tools to actually manage the site. Figure that'll give me the best feeling for what folks who don't just "pop into the shell" go through. So, I used Fantastico on my hosting provider to install Wordpress, even though that tends to keep you one or two versions back from the latest and greatest, since after a new version is released, the hosting provider has to try it, then after they try it, they upgrade the information for Fantastico to install. Well, I'd waited long enough that Fantastico was sending me "you need to upgrade" mail. OK, fine, I'll do it.

No matter which route you go, however, upgrading things can be a challenge. I use Wordpress for my personal blog here at Kencasting, and have done a small set of customizations. Put them all in a separate theme to keep them easy to keep track of. And, I have a small set of Wordpress plugins as well. Nothing wild, customization-wise, pretty minor, actually. But, I went to the wordpress upgrade page and went through all the steps, making sure I had plenty of backup in case something went wrong.

So, it took me about 40 minutes, and I did pop into the shell to tar up some files, since it was just faster to back them up on the server, rather than ftping them off the site, then putting them back if I needed to. For the rest, I used the online SQL backup tools (phpMyAdmin was provided by Bluehost, but was glad I had the step by step to walk me through it, too many options, really needed a basic "backup all my stuff" route).

This really drove home again the potential benefit of "appliance-izing" (or should it be appliantizing?) various common web-based tools, who's management should be offered more or less as a service. I don't want to giving up *all* management by using a blogging host. A software appliance version of Wordpress, with built-in management tools, could do a "backup and try it" in one step, checking things like plugins for known good (or bad) versions (just like Firefox does for me all the time), and should be able to keep track of what I've changed, so I don't have to remember. After all, I'm jumping from "editing php in a management window" to "now backup those files", which I can navigate, but I'm an old Unix hack. I shouldn't have to know what I had to know to simply upgrade my blog. That'd make the basic process much easier. And, that's just something relatively simple, what if I also wanted to manage the a variety of other hosted apps in the same way? Then it gets much more compelling.




Powered by ScribeFire.

http://blogs.sun.com/wallich/date/20070607 Thursday June 07, 2007

A great article

Marc Andreessen wrote a great article on why there's no such thing as Web 2.0. I totally agree with him, very well written.

Thanks to Oren for the pointer!


Powered by ScribeFire.

http://blogs.sun.com/wallich/date/20070522 Tuesday May 22, 2007

Not so rambly

Well, haven't been that rambly here lately. Still working out details, sketching out prototype ideas, and starting up the team.

But, figured I'd just check in and quick a quick update. To tide all my loyal readers over, here's a little fire and lightning from this last weekend's Maker Faire. Although not exactly your normal kind of technical conference (it's a faire, after all), I did meet up with some folks I knew, and got some ideas for projects to do in my spare time.




Powered by ScribeFire.

http://blogs.sun.com/wallich/date/20070426 Thursday April 26, 2007

Virtual appliances on my Mac

Just installed VMWare Fusion and Parallels Desktop on my Mac. Wanted to start mucking with the available virtual images out there, and start hand-crafting my own to get a feeling for how to not only make and distribute images, but use the tools that can convert between them. One such tutorial can be found at Virtualization Daily. Another option from Parallels is Transporter.

I'll be putting together (and chronicling the process here) of migrating between the different virtual disk formats, and trying some of the different appliances.

My first impressions are that Parallels seems to bog my Mac down less than VMWare, but that's on a 2Ghz MacBook with 1Gig of memory, 512Meg for the virtual machines. When running just the image, with everything else shut off, they both perform pretty well, and almost comically better than Virtual PC did on my dual processor G5. That's after turning Fusion's debugging off (it's on by default). When on, it was really sluggish, and I wasn't very happy with the results.

Parallels has some cooler look-and-feel bits, but VMware has the feeling of having done this before, and knowing what you'll need.

One totally unscientific test, I opened a Jumpbox mediawiki image I had in both Parallels and VMware format, the Parallels one came up about 30 seconds faster than the VMware one. But, it's not fair or scientific, since I don't have enough memory for them both to run without some sort of interference yet.

More later, just a quick note as to what I'm experimenting with.


Powered by ScribeFire.

http://blogs.sun.com/wallich/date/20070405 Thursday April 05, 2007

Legos

As I talk to more people here at Sun, and nail down what'll add the most short-to-medium term benefit, as well as long-term benefit from me, I'm getting closer to a crisp roadmap.

On the way to that roadmap, I wanted to share an analogy of what I'm working towards here.

Legos.

Sun's been pretty good at shipping boxes of legos. Sometimes the boxes contain sub assemblies of legos too. They often come with an instruction sheet for building various things out of the legos, to get folks started. Legos are great, with a box of them you can build a geometric shape, a car, a robot, or a pen plotter.

But, there's also a market for pre-built blocks, cars, robots, and pen plotters. Not everyone wants to build their object. Some just want to use the resulting object that will get built, and don't have the need to customize it in ways that only the builder can do.

A term for special-purposed devices has been being used for a while, especially as it applies to software and hardware combined for a specific purpose, an Appliance. NAS appliances are a well known version of this, in fact, a group at Sun is building one of these as well. I have a consumer one at home (my personal blog details the long story of that). It does one thing, provide accessible redundant storage. It does it efficiently and quickly, in a small attractive enclosure. It cannot, however, run general purpose applications to, say, backup my iPod. Why would it? It's a NAS box.

Other obvious examples, most home firewall/routers. They've got an operating system (often a Linux derivative), and management screens. They've got a web server running on them, and sometimes a mail server. But, can I run OpenOffice on it? I think not, it's a router.

Now, there are also examples of appliance boxes that are useful, but limited by various factors. They're a playground for hobbyists. TiVo and AppleTV come to mind. But, all you have to do is check out Make magazine to find out that any ordinary appliance to most people is a playground for customization by others.

OK, so other terms seen about lately are "Virtual Appliances", and "Software Appliances" (IDC even has a definition for how the two are similar and different). In a nutshell, a virtual appliance is some set of software, wrapped in packaging that will run one one of the several virtualization platforms available: VMware, Xen, etc. A software appliance can be a virtual appliance, but it could run on bare hardware as well. This may be another way of expressing the thought here, although at present, the whole Appliance name is both overloaded, and under-defined.

What makes a Software Appliance different from a software distribution? I was at a Virtual Appliances Leadership Summit, hosted by rPath with members of many virtualization, hardware, and software companies, discussing many of these issues this week, and the answer is, well, folks are still defining their version of it.

Here's my quick and dirty definition. An appliance must, out of the box, with little to no work by the user, serve one or more functions. In addition, it must be easy, actually trivial, to:

  1. Install
  2. Use
  3. Maintain
  4. Support
By definition, it's not a general purpose machine, so you don't plug an install DVD into it and put more things on. It's also not endlessly customizable, we already have plenty of ways to build those systems, if that's your goal.

So, to end today's ramble, I'll leave you with this. My goal here at Sun is to take our current sets of Legos and sub assemblies, and provide a few cars and robots, a way for people to obtain and use them, and a way for customers who want these kinds of object, to create their own cars and robots. For those at Sun, I'll be doing that by utilizing what we have, not reinventing things we already do. If you're doing something that sounds like it fits it, and we haven't talked yet, give me a shout. We'll still be shipping the subsystems, that's our bread and butter today, but as software evolves to a service on top of which new companies add value (i.e. YouTube, PodShow, Fuzz), there will be an increasing market for turnkey, easily administered and supported appliance-like things.





Powered by ScribeFire.

http://blogs.sun.com/wallich/date/20070315 Thursday March 15, 2007

Week 3

Finishing up my 3rd week. Quite the dip in the deep end, but it's still quite exciting. Focusing on how to take what we've got in Software, make sense of it, find out who's doing what, and what areas people really needed help in solving.

My particular challenge is how to create turnkey software stacks, with a sub-challenge being to help define what pieces are needed for those stacks, and what pieces are available internally, and which of those are being made available externally, and how.

My other goals is to start "working from the end state", what problems are being solved, and why, to determine the appropriate toolbox needed to solve those particular problems. As an engineer, it's all to easy to look in today's toolbox, and figure out exactly how to solve any problem with the tools at hand. That's great when you're delivering to other people just like you, but if you're delivering (directly or indirectly) to someone who isn't like you, you've gotta figure out how to step back.

More on all that in the coming weeks as I nail down the how, now that I'm getting a handle on the what.


Powered by ScribeFire.