Preview: Jef"I am the pusher robot"Spaleta
Jef"I am the pusher robot"Spaleta
Jef"I am the pusher robot"Spaleta - LiveJournal.com
Last Build Date: Fri, 02 Mar 2012 19:45:17 GMT
Raspberry Pi. and what it means for boutique electronics more generally.
Fri, 02 Mar 2012 19:45:17 GMT
Anyone out there actually get lucky enough to buy one before the distributors sites failed in the initial onslaught? I didn't.
I've had to make due with reading the laypress reaction articles.
The interview with Mr. Upton concerning the project history is very informative.http://www.linuxuser.co.uk/features/raspberry-pi-interview-eban-upton-reveals-all/
In particular his answer to why the project is incorporated as a non-profit is very interesting. Especially the bit about how component distribution channels directly impact the cost-control for niche products (10k units is niche for the purposes of this discussion).
He points to a subtle but important problem in device innovation at present. The distribution channels are geared to hinder boutique or small-batch crafted innovations from being cost competitive with mass produced goods. This isn't about volume discounts. We all understand that boutique and small batch goods have higher costs associted with their higher value. I prefer locally produced microbrew beer for a variety of reasons even though Budlight cost so much less. No, this is about the unnecessary operational markup imposed by the distribution channel model component manufacturers force small scale for-profit innonvators to use.
Even though the components being used are all commodity..it is the inefficiencies of the distribution channels which place the overhead of middelmen distributors between the product producers (the chip manufacturers) and the product consumer (in this case the raspberry pi project team) The fact that a small non-profit can arrange a direct relationship for a small volume order with component producers in a way that a small for-profit startup could not is a problem. Being non or for profit shouldn't matter. There's absolutely no reason for the arbitrary difference in access to small volume orders at a fair price. The prices raspberry pi are not a sweetheart deal..Mr. Upton confirms that explicitly...its just a policy decision to ignore small volume for-profit customers and make them deal with the markup of distributor middlemen. It's like telling my local microbrewers they could avoid using a hops wholesaler just because they were a non-profit.
In an age of 3d printing and on-demand self-services...electronic component makers need to do better. They need to find a way to service the boutique electronic industry in a much fairer fashion and learn to leverage them as part of an innovation cycle that lowers production costs for the the entire ecosystem of boutique goods. I want an affordable microbrew economy for electronics that can sustain local companies everywhere. I don't want a few large scale distributors of finished goods being the only viable for profit companies who can compete for consumer dollars.
My obligatory GNOME3 post.
Thu, 16 Jun 2011 12:01:33 GMT
You know what I could really really use.... GNOME shell keyboard shortcut stickers for my keyboard.
Little key shaped, mostly transparent so I can still read the key underneath, stickers that give me some hinting as to what keys did what in combination with alt clt and super.
I thought of this when I was using my wife's desktop the other day and she has put the Gmail keyboard shortcut stickers (provided by Google at one point) on her keyboard as a learning tool.
I'm never going to remember all the useful shortcuts exposed in Gnome shell just by reading documentation. Not going to happen. And we've got no hinting in the UI itself for a lot of these things at the moment. And that isn't a knock against GNOME3 specifically...its the nature of the beast. Applications with drop down menus will try to give you hints about accelerators..but there's a lot of stuff that predates G3 Shell where the keyboard interactions aren't learnable through use. If the GNOME designers find a way to solve that problem, then bonus points for them. In the meantime, training stickers on my keyboard are the best idea I can come up with for my own needs.
Now I can of course do up an inkscape svg template for these things and share it with people. But if anyone else is already looking at doing this sort of thing...let me know.
Ode to the Cloud
Tue, 14 Jun 2011 15:20:33 GMT
Inspired by Spot's SELF 2011 keynote, I got drunk and penned the following parody lyrics to the infamous "Log Song" from the old "Ren and Stimpy" cartoon. Those of you who know it, feel free to hum along.
What costs pennies in fares alone or in pairs?
A fav with the geeky crowd?
What's great for a stack and sits in Amazon's rack?
It's Cloud, Cloud, Cloud!
It's Cloud, Cloud....
It's awfully misunderstood....
It's Cloud, Cloud...
It's better than bad, it's good!
Everyone wants a Cloud! You're gonna love it, Cloud!
Come on and get your Cloud! Everyone needs a Cloud!
Cloud! by BLAM-O!
-jef"Cloud Cloud Cloud Cloud"spaleta
Looks like I'm going to SELF
Mon, 02 May 2011 17:00:18 GMT
Due to a set of very interesting circumstances I'm going to be flying from Alaska back to the East Coast for a couple of weeks. And apparently the plane ticket fairy has decided to make it possible for me to get back down to Charlotte on condition that I show up to SELF down in Spartenburg and not just spend every day commuting between the _new_ Charlotte curling club and the whitewater park. Note to myself: remember to bring the correct paperwork(including my long form NC birth certificate) to let me cross the border between North and South Carolina. I've been in Alaska so long I've forgetten what the rules are for interstate travel.
I hope there's some sort of door prize at SELF for the attendee who has traveled farthest.
A revelation of an idea for GNOME3's Shell
Sun, 24 Apr 2011 21:10:34 GMT
I've made heavy use of the revelation-applet on my gnome 2 desktop panel to search for passwords inside my revelation password file. I'd like to restore that sort of quick password search inside the preferred mechanics of GNOME3 shell.
So here's my thought. In the activities overlay would it be appropriate to extend the search box so that I can search an encrypted password store? I don't really care if its revelation concept of an encypted store or something else but it seems to me the mechanics of the search interface in the activities overlay is providing the same search mechanics I was already using...I just need to be able to point the search at a _local_ password store instead of wikipedia or google.
Intuitive, You keep using that word...
Fri, 22 Apr 2011 19:35:05 GMT
"You keep using that word. I do not think it means what you think it means" - The Princess Bride
Of, relating to, or arising from intuition.
The act or faculty of knowing or sensing without the use of rational processes; immediate cognition
Technologist, and the culture they are embedded, seem to be trying to redefine what intuitive has historically meant. I'll grant you that language is always evolving and the shift in this particular definition is just part of that evolution regardless of what some silly dictionary would have you believe. But I think it does us all some good if we stop and think about the traditional sense of these words and make sure we aren't falling into a trap of wishful thinking about what we really wished others would define them to be.
Something that is intuitive is not inherently rational or derived from a rational systematic thought processes. Intuition is not deductive reasoning.. is not the synthesis of observable stimuli or facts into a coherent mental model which can then be used to make predictions. Sherlock Holmes and his latterday counterpart Monk aren't fictional masters of intuition, they are characterizations of extremely skilled deductive reasoning which is the exact opposite of intuition. Attention to detail and reliance on methodology in order to make something complicated look easy, doesn't require a single ounce of intuition.
Intuition is not teachable. It is not a process of discernment which can be followed. Intuition is at best situational and at its very worst a very individual specific trait that defies quantization. Intuition and its intuitive fruits are the stuff of dreams and nightmares...of genius and madness.
So when people describe a technology or an interface as "intuitive" keep in mind these traditional definitions about what intuition has historically meant instead of just redefining it as "easy to use" and we all just might communicate a little more clearly. Perhaps in reality building things which rely on a person's "intuition" to understand how they work is ultimately the wrong approach for all technology. Perhaps if we build technology and systems which rely inherently on "deductive reasoning" or other teachable systematic way to take in new information and synthesize it into a mental model is a far better way to proceed.
Words can matter. So can methodology.
Let me take a moment and set some expectations.
Fri, 15 Apr 2011 18:43:03 GMT
If you have ever in the past mistakenly assumed
that I work and speak for a company that competes with your employer because I have been vocal in expressing my personal opinions which just happen to be critical of your employer....
And then you decide to claim it is offensive for me to even question
whether your public speech might be influenced by the interests of your current employer....
That earns you a place on my growing list of people who I'll go out of my way to be critical of as an individual who may or may not have unstated motivations coloring their statements. I just want to be clear of that. I will attack
this sort of hypocrisy unapologetically.
I don't have a problem with people making the mistake of assuming that I have unspoken intentions and motivations. I would prefer those people to ask me to clarify instead of assume. But hey I know the score, everyone makes assumptions. I can forgive those human
But what I cannot forgive is the hypocrisy
of making those assumptions selectively simply because I disagree with you and not applying the same assumption filter to yourself. So for the rest of you out there who may read this, and who may find themselves stuck in a discussion with me over some sort of heated topic. Considering this your warning. You want me to play nice? You want me to go out of my way to avoid offending your very delicate sensibilities? You want me to be humane? Then do yourself a favor, don't blithely assume I'm speaking as the shill for a corporation I don't even work for.
This goes double-so for managers
at corporate entities who I might engage in public discussion over corporate policies that I disagree with. Think very carefully about what it says about your own corporate culture when you find yourself assuming I am speaking on behalf of a corporate entity I don't even work for.
Public discourse is important.
Mon, 07 Mar 2011 16:31:50 GMT
When you or your employees make vague statements about being prejudiced against in the activities of a multi-vendor project, and those allegations are being made in public forums...but you and your employees refuse to provide evidence of those allegations in those public forums... I am not amused.
And when I publicly challenge you to provide such evidence and you give me information in private email but do not give me permission to rebroadcast that information... I am angered. This sort of back channel "persuasion" does not get to the heart of the problem. Because well, I am not the heart of the problem. I may be your PR problem..but that's not the real problem. Convincing me of the righteousness of your opinion doesn't help solve the underlying problem. If your hurt feelings are substantiated by the evidence, having that information stop with me..and preventing me using that evidence to start _fixing_ the actual problem is not in your best interests. When such back channel information sharing is occurs I view it as manipulative.
I don't have a problem if someone takes a discussion with me private because they feel they need the space that a private email discussion provides to focus without sidebar comments or commentary. But when the discussion is over I fully expect to be able to republish a private conversation in its entirety. It's only through publicly disclosed discourse that people can be held accountable for what they say, including myself.
Wed, 02 Mar 2011 05:02:15 GMT
So a little while back I wrote a blog with a few questions about the future of LibreOffice and what part of its restructuring is going to matter the most.
Well, here is Micheal Meeks talking about the progress with Libreoffice.http://www.youtube.com/watch?v=-Olx3EvJMl0
The first think I take away from this presentation is that LibreOffice's restructuring is "proof positive" that copyright assignment policies _hurt_ projects.
It's also got a good discussion of the benefits of a community project to do things that a commercially backed development can't easily do.
I also like the bit about the local minimum in doing open community development. You get the benefits when you go all the way..but there is a middle point where you hold too much control and you get none of the benefits of open development even though you are trying to do it.
Very very fascinating talk.
Corporate ethics is grounded on accountability to the public.
Mon, 28 Feb 2011 06:39:09 GMT
This post is about Canonical and the Banshee situation that has unfolded over the last week or so. Specifically I am going to address the point that has been raised concerning whether or not Canonical has an ethical obligation that goes beyond what is allowed by the licensing associated with Banshee as a FOSS codebase.As a corporation Canonical cannot expect their actions to just meet the minimal standard of legality. As a corporation Canonical must adhere to an ethical standard set by their consumers, their business partners and the community the rely on. Throughout history corporations have been successfully pressured by the public to change their behaviour when the public have found such behaviour to be undesirable and or damaging for the social good. There are multiple examples over the last 20 years of boycotts and protests where the public have coerced a behaviour change from corporate entities above and beyond what is strictly legally required. Canonical must keep this in mind when choosing to craft each and every business decisions it makes. Corporate accountability ultimately rests in the court of public opinion. Sometime that public opinion gets enshrined as a new legal requirement when corporate abuses are extreme. But many other times, accountability and corrective action comes simply from people speaking out (respectfully) when they see a corporation doing something damaging.This is no different than the historic example of people protesting Nike for using foreign “sweatshops” to produce their products. That protest spanned a decade, with Nike continuously being defensive about their policy and saying it was perfectly legal. But eventually Nike smartened up and changed their corporate culture and addressed the “unethical” business activity.http://en.wikipedia.org/wiki/Nike_sweatshops“Nike was heavily criticized for selling goods produced in sweatshops throughout the 1990s. They originally responded by lashing out and denying all claims brought against them. However, Nike’s director of compliance, Tom McKean, spoke of Nike’s irresponsibility in 2001. McKean stated in an interview that, “Our initial attitude was, ‘Hey, we don’t own the factories. We don’t control what goes on there.’ Quite frankly, that was a sort of irresponsible way to approach this. We had people there every day looking at quality. Clearly, we had leverage and responsibility with certain parts of the business, so why not others?” Recently Nike has developed an intense program to deal with these claims. Nike has employed a staff of 97 people to randomly inspect several hundred of their factories each year. Nike also gave the Fair Labor Association, an association founded to monitor labor conditions, the privilege to randomly inspect any factory they wish.”While the Nike analogy is not perfect(no analogy is) its stresses the point I want to make. Perfectly legal business decisions can be irresponsible and can be damaging to both your corporate interests and to society at large. Whether a business decision is legal or not is the absolutely lowest bar by which a business decision can be evaluated, it is never enough to define a corporate culture and a corporate brand that people respect, it will never be enough.Where Nike built its image with consumers on quality and eventually learned to internally that focus on quality inside its own business practises, Canonical needs to learn to internalize the respectful ethical engagement with community that it projects in the minds of its consumers. Canonical built its image..builds its products..based on a standard of ethical community engagement it set for itself. It is unreasonable for Canonical expect that it won’t have the Banshee interaction held to the same standard of community engagement that is used elsewhere inside [...]
Tue, 07 Dec 2010 01:54:27 GMT
Is it respectful to actively encourage other people to contribute code to a project by pointing them to a set of getting involved instructions for that project which encourages them to write patches and code without making any mention at all that a copyright assignement is required before those patches and code contributions can be merged into the mainline development branch?
I don't think that is very respectful. It's certainly not intellectually honest. Personally I would think you would want to be as upfront about legal requirements for contribution as you can in an effort to be as honest and transparent with the contributors you are recruiting to help. It would seem far worse to me to have someone spend time on code and then tell them after they have produced it that they must sign a contract before it can be included. That seems like a manipulative strong arming tactic to me. Get the contributor to do the work and then tell them about the legal strings attached to having the work be considered for inclusion.
If we are going to have a larger conversation about what respect in the FOSS ecosystem looks like, we need to make sure that conversation includes the topic of what is fair and equitable notice with regard to legal requirements when recruiting contributors.
Thoughts on LibreOffice.
Tue, 28 Sep 2010 20:34:18 GMT
1) Will LibreOffice gather a critical mass of developers and push the codebase forward faster than the rate of development of OO.org? If yes, can we quantify which of the project structural changes make that possible?
1.1) The importance of independent foundation control?
1.2) The importance of the lack of a copyright assignment requirement for contributors?
1.3) The importance of independently run infrastructure as a basis for development?
2) Assuming the contributor critical mass is achieved, can the Document Foundation use LibreOffice as a starting of a codebase that provides an open, replicable (possible decentralized) collaborative internet-centric document services that can be integrated with or used as an alternative to centralized online document store services such as Google Docs.
The netbook is dead....
Sun, 11 Jul 2010 04:35:56 GMT
or so some people would have you believe.http://news.cnet.com/8301-13924_3-20007756-64.html
Apple's Ipad has apparently taken a chunk out out of the netbook market with its introduction. I could nitpick the numbers a bit and say the jury is out still on that, but for the purposes of this post I'll play along with the analysis and accept that yes the Ipad did take a bite out of netbooks sales.
And if you follow the logic the implication is that "slates" as a form factor will continue this trend. I'm not sure I believe that, I'm not convince that another OEM can ride Apple's coat tail with a completely different "slate" form factor offering. I certainly think the netbook formfactor as a market doesn't have a good identity and people are making the best of fitting them into their lives because they are just so darn inexpensive. The ipad...not so cheap..but its eating into netbook marketshare. There's definitely something very important going on there..but it may not be easy for another OEM to duplicate. They are going to give it a good college try however.
HP now has its own in house mobile platform in WebOS with the purchase of Palm and seems to be saying that its going to be the basis of its own slate offerings:http://www.wired.com/gadgetlab/2010/05/official-hp-slate-will-run-webos/
And all indications seem to point to Dell's Streak running Android.
All of these operating environments are first and foremost mobile device environments. Not traditional "desktop" environments. And while webOS and Android are based on linux, neither of them or traditional linux "distributions." I think that is very interesting as a contrast to the evolution of common netbook environment. Netbooks are essentially cutdown laptops and the traditional linux distribution model mapped over without much fuss. Even the linux arm netbooks which are due to sweep across the globe like a pandemic.... any day now.....they take a little more work because of the change of architecture but that the end of the day its the same user environment.
Are traditional linux distributions ready for the slate form factor? Are traditional linux distributions even going to be possible on these devices? Unlike netbooks, I think slates as a market segment are more inclined to fall into a pattern of OEM enforced walled application gardens...with 10 foot tall walls in the form of locked down OEM pre-installs. Sure all the android devices may have access to similar apps. And all the webOS devices may have access to the same apps. But cross environment applications maybe far rarer than what we are use to in traditional linux software distribution ecosystem. Where Fedora, Debian, and Gentoo share much of the application-scape across their boundaries.. those boundaries are essentially decorative garden edging.. not wall that have to be scaled by application developers..which for the most part rely on QT/KDE or qtk/gnome frameworks to build applications that should work on any of those distributions without significant distribution specific hacks. I get the feeling that's not really the same for webOS and Android.
And with all the work going on right now to support netbook oriented interfaces in the more traditional linux software ecosystem..having a market research firm stand up and say slates are going to kill netbooks doesn't feel so good.
How important is copyright respect for a healthy FLOSS ecosystem?
Fri, 21 May 2010 00:11:14 GMT
I have a very hard time understanding how multiple outspoken members of a FLOSS community can find it socially acceptable to lift up creative multimedia works made in their community, meant to show off what is great about their community, but are themselves flagrant infringements of the copyrights of others who are doing their work outside of the FLOSS community.
Isn't respect for copyright and the rights given to content creators to set the terms of usage on their works a cornerstone of how a vibrant FLOSS ecosystem has been able to self-organize? FLOSS and Creative Commons licensing is a choice that content creators make.. and when they make choices about licensing which are contrary to the ideals of FLOSS and open media its entirely inappropriate to disregard those choices and to incorporate their code or their content as an element of your own work just because its available for easy consumption. I don't care if that is code or documentation or still photos..or music.
Taking a song you legally purchased online for personal use and shoving into your video you plan to redistribute widely as background music is simply disrespectful of copyright and actually undermines the open media movement. And frankly when that video is meant to showcase how great your FLOSS community is.. is both ironic and damaging. I don't care how great the song is. I don't care how cool the video is... I don't care about how open and friendly your community is...the ends don't justify the means. You can't build an argument to have people respect FLOSS and open media licensing while at the same time completely disregarding the copyright licensing of mainstream popular music.
Thu, 25 Feb 2010 00:04:21 GMT
For the sake of argument, I'll accept, for the moment, that software patents as a concept can be effective as limited monopolies granted inventors as a means to promote the progress of science and useful arts.
Accepting that, what continues to baffle me is why its okay to make statements which imply patent infringement in a piece of software, to exhort money via private settlement agreement from users of that software, but then to deny the original creators of that software the ability to be told in specificity what the infringement claim is so that it can be addressed.
It's one thing to have a system of patents laws that attempts to see an equitable share of profits back to the original inventors via an infringement resolution process (whether it be in a court of law or via private settlement.) That is right and proper. It's quite another to have a system that actively encourages continued infringement as a revenue stream because the rights holder never has to reveal the infringement claims to the actual creators of the infringing work and in fact makes it a condition of settlements with users of that work that the original creators of the infringing work cannot be told what the specific infringement claims are.
Infringement should not be a revenue model to be managed and milked, by deliberately picking out technology users who are most likely to settle quietly but who do not control whether the underlying technology at issue is infringing. If there is an infringement claim, even if its settled out of court and behind closed doors, the creators of the original work should have a right to know what the claim is and should have the right to attempt to address the underlying infringement for all future users by re-engineering their software. If you are going to sue someone for using the linux kernel because it infringes your software patents, then you should be forced to tell the linux kernel developers what the infringement claims are.
Android, the most popular linux distribution?
Mon, 22 Feb 2010 03:10:51 GMT
Apparently nearly 8 million Android equipped smartphones shipped in 2009 
. And more recently in 2010 Google's CEO has been quoted as saying that there are 60,000 Android devices shipping per day (and that the rate is accelerating)
. If this trend holds, doesn't this make Android the most popular linux distribution (and its not even listed on distrowatch)
Audrey versus Litl: What a difference a decade can make.
Fri, 18 Dec 2009 05:07:05 GMT
I think the 3.com Audrey was the first computing device I remember that was meant to be strictly a home internet appliance device.... back when dial-up was the cornerstone of residential connectivity and VCRs were the dominate home entertainment format.
I think its interesting to compare the design decisions that went into the Audrey and the decisions that went into the Litl. The biggest difference is of course the Litl's focus on consumable digital media....something that didn't really exist a decade ago. And of course Litl's wireless support..again something that wasn't common a decade ago. But beyond that there are a lot of similarities. The both have the concept of channels. Audrey had a knob.. litl has a wheel. Audrey had a wireless keyboard... you can hide. Litl's keyboard hides in easel mode.
What I really want to know is how hack-friendly is the Litl device? Audrey was extremely hack friendly and as a result the hardware itself remained quite usable beyond the very short lifetime when its online services were available. A loyal cult following of technically proficienty Audrey owners were able to spin up multiple alternative QNX images to extend Audrey's functionality well past the original designers intent (and well past 3.com's interest in selling and supporting the device.)
Fun facts..... about copyrights.
Wed, 16 Dec 2009 21:14:34 GMT
If you decide to put a cover band together and have a little jam session at a computer software conference or a frat party in Dallas, Texas and perform a Willie Nelson song such as "On the Road Again" for a group of conference attendees... that is a public performance and is you should be paying royalties. If you also decide to record that performance and make it available as a digital download from your employer's corporate servers... you should be paying another set of royalties per digital download.
If you are an employee of a software company that plans to open up its own music store in the near future, and have done all this without getting the permission of the copyright holder nor have paid the statutory royalty amounts for neither public performance nor digital distribution... that's just tragically ironic. If those digital files are hosted on corporate servers... that's doubly ironic and opens up the employer to potential copyright infringement liability.
Can corporations be friendly?
Mon, 14 Dec 2009 18:44:34 GMT
Inside the current crop of blog posts like this one about the MySQL/ Oracle acquisition.. there is insight into a deeper issue which I want to talk about.What does it mean for a company to be friendly or unfriendly? Can a friendly company make a certain set of licensing choices and be applauded while an unfriendly company makes the exact same licensing choices and is derided? This inconsistency is not rational and we need to understand why some in the community are reacting this way.I think what's happened in recent times is that as a community we've had a lot of start-up business interests that are in effect cults of personality. One or two dedicated, passionate people building a company around their personal ethos. For these small incubator companies... the founder is the company.. there is not self-organizing corporate culture. The company is thought to be friendly because the founder is thought to be friendly. But its the wrong way to view a corporation. Corporations aren't people. Corporations are for all intent and purposes...alien..to the human condition.The cult of personality model only really holds for very small companies...where the original founder is still in place. It does not hold for companies that outlive their original leader. At some point companies develop their own corporate culture..which is an amalgamation of choices being made by multiple people..and not just the founder. At some point companies stop being cults of personality and take on a life of their own. And in this sense I think Red Hat and Mozilla stand apart. They've made the leap into self-motivated corporate organism and have retained an open development friendly business culture.. and they've taken vastly different paths to get there. But as business cultures aren't static, in time unfriendly cultures can turn friendly and vice versa...because they are an amalgamation of choices. We have to do our part to lift up the good decisions inside a corporate entity that are being made while at the same time punching them in the head for making bad choices. We can not paint large corporate entities broadly as friendly or unfriendly...Intel and Google I'm looking at you. The MySQL situation is a cautionary tale for every single open-core start-up in existence right now. Every single company which requires copyright assignment from contributors and is making a business off of owning the copyrights by holding a privileged position that allows them to license the code under both proprietary and open licenses. The people who are upset with the MySQL/Oracle situation should be looking to the future at what's going to happen to all the start-ups out there who are following MySQL's licensing model. What if Alfreso is acquired? What if Canonical is acquired? Are we going to go through the same sort of debate every time an acquisition that involves privileged dual-licensing happens? Why exactly as a community are we cool with signing over copyright rights to small corporate entities.. when we know that if they are successful, acquisition by a larger corporate entity is a probable outcome? If we don't want to give Oracle or other large "unfriendly" licensing control over a codebase why the hell are we signing over copyrights over to small companies like MySQL? Anyone who thinks Oracle's sole control over the MySQL codebase is a problem should have foreseen this and spoken out against the open-core licensing model when MySQL adopted it as a business strategy. Whose going to start demanding Alfresco stop requiring copyright assignment now? -jef[...]
Will the real statistics junkie please standup
Wed, 04 Nov 2009 08:26:13 GMT
This is in response to http://nicolas.barcet.com/drupal/en/oct-ubuntu-server-statsTo get details stats on OS breakdown from netcraft to put that 1.4 million Ubuntu web servers in the correct context... you have to purchase their data product. Problem is... they also restrict how you can use that data so even if Canonical purchased it for you to look over..you probably couldn't comment on it publicly. The full Netcraft survey data is problematic in that regard because you really can't have a public discussion. But we can have a useful discussion about the 2009 Purchasing Survey because they publish their methodology AND they raw survey data.The purchasing survey is a really mixed bag of news when you read the whole article and look at the raw data. The article really begs the question... where is that stated growth of Ubuntu server deployments coming from? The article specifically makes the claim that windows to linux migrations are stalling and that people are less likely to dump windows for linux. So where's the Ubuntu serve deployment growth being generated? Virtualization maybe? Not according to the raw survey results.If you dig into the raw data...you'll see that exactly one survery respondent(out of 459) said they were using Ubuntu/Debian based KVM for virtualization. And more sobering only one respondent (out of 449) said they planned to deploy Ubuntu/Debian based KVM in the next 12 months. That should raise some eyebrows inside the Canonical fenceline. Doesn't that survey result run counter to pretty much everything Canonical and its virtualization partners have been saying? Hopefully they'll repeat these virtualization usage and intent to deploy questions in next year's survey after the next Ubuntu LTS is out and both Canonical and Eucalyptus Systems are pushing Ubuntu server for private deployments. But more generally speaking I'm not sure that the Purchasing Survey results are self-consistent enough to be reliable. For example look at questions 22 and 67.question 22: Which server operating systems do you currently have installed? (Select all that apply.)question 67: Which of the following Linux distributions/operating systems do you currently use on your servers? (Select all that apply.)The numbers don't compare well across those two questions. There is at best a 10% point discrepancy in the Red Hat deployment percentages between those two questions. There is a similar discrepancy in the CentOS numbers. That's not a good sign for survey accuracy. If there really is a 10% point error, that potentially wipes out the implied Ubuntu growth in the summary article. And I'm not saying that the Ubuntu growth does not exist. What I am saying is that when you look really closely at the survey data.. the survey does not appear to be accurate enough to say anything statistically significant about Ubuntu growth if the noise floor in the survey really is 10%. The survey summary article consistently overreaches in its conclusions without once commenting on the inherent accuracy limitation of their survey. The point I'm trying to make...to everyone.. is that you can't just throw numbers up without considering the accuracy of the methodology. For this survey in particular... If they can't get Red Hat deployment numbers accurate to 10% between question 22. and 67..the linux distribution with the most respondents and therefore the best statistical accuracy...then you can't really expect the other linux [...]
The age of E-Ink.
Sat, 24 Oct 2009 20:50:23 GMT
Good news on the E-Ink front.
I personally think that E-Ink is one of the biggest technology innovations in like the last decade. It's been slow to be adopted since its public debut... but I really think its an important technology that has applications well beyond E-books.
Android and MIPS sitting in a tree....
Thu, 01 Oct 2009 17:22:09 GMT
Well, maybe not a tree... but sitting in your settop cable box or integrated into your HD TV or bluray player...maybe.
ARM/Android gets a lot of attention from the laypress right now, but we don't hear a lot about MIPS. But if you take a real close look it seems there's a very coherent strategy in place to position Android as an ubiquitious device OS for MIPS based devices Once you add MIPS into the picture, it sort of makes sense how Google can want to spin up both ChromeOS and Android as operating systems without stepping all over each others toes.
Whats fascinating about MIPS is how absolutely under the radar it is as a technology..even more so than ARM. Do you know how many pieces of electronics you own that has a MIPS chip buried in it? Android is going to "win" the post-desktop device landscape and noone is really going to notice it happening, just like noone really notices how many "computers" they really interact on a dialy basis. I'm pretty sure that when Google does finally unveil the Google neural implant ( will that make us all androids in a sense?) that there's a good chance a MIPS chip will be inside it.
The cadence of discussion
Tue, 29 Sep 2009 19:29:21 GMT
Can anyone point me to any upstream project (that does not require copyright assignment to Canonical) that has picked up the meta-cadence idea and as a project trying to define something like a 2 year meta-cycle on top of a faster moving time based release cycle? Shuttleworth has been talking about meta-cycles for over a year now and I haven't seen any public discussion which indicates that upstream projects are warming up to the idea of a meta cycle at all. Have I missed something?
Banned for discussing the ethics of screensavers
Tue, 22 Sep 2009 17:48:57 GMT
So campers.... what do you think. Is it a good idea to run screensavers that scrapes random images and text from the internet? I think its a bad idea. I think it was a bad idea when xscreensaver did it with rss feeds and web images and I think its a bad idea now. We should not be encouraging the pulling of random content from the web to a screensaver. Content has context..and there needs to be a human being sitting in front of the display making a judgement as to whether the content is appropriate for viewing at that moment. Having unattended screensavers grabbing random material is a garunteed way to display the wrong content in the wrong context to the wrong people.
This maybe cute functionality to code up...and its certaintly trivial to implement. But just because its cute and doable doesn't mean we should encourage developers to do it and make it available to people to use. Don't developers have some responsibility in choosing to refrain from implementing functionality that could have problematic social impact? I think they do. And I think people who lift up this sort of problematic functionality also bear some responsbility...especially when they don't remark on the problematic nature of display random content in unmonitored situations.
Appearently a certain Canonical employee doesn't feel its appropriate for me to express these opinions as part of a discussion in his otherwise open blog comments even after the actual developer of the software chimed in and essentially agreed that my concern was valid while disagreeing with me that as a developer he has some responsbility for choosing to implement inherently bad functionality. http://castrojo.wordpress.com/2009/09/15/announcing-lifesaver/
Hey if that's what Canonical wants to see... random profanity ripped from twitter and identi.ca in Ubuntu screensavers activated in public spaces... more power to them. Say the word, and I'll gladly help accomplish that by encouraging wide usage of this screensaver. In fact I'll go out right now and boot live version of Ubuntu on with the lifesaver screensaver activated on all the public computers I can find since Jorge, a Canonical employee feels its a good idea.