Monthly Archives: August 2006

My Zaurus C3200 Linux PDA

By  | 

I thought that I really wanted something significantly smaller than my four-year old laptop but with a hard drive and more functionality than a standard PDA. After a few weeks of research and reading articles about different devices, I decided to buy a Sharp Zaurus SL-C3200. Strangely, these are not sold in North America so it required ordering it through the web, site unseen. Why choose the C3200? First, it comes preinstalled with Linux, which was one of my requirements. Second, it has a hard drive. I don’t understand why I can purchase miniscule MP3 player devices with 30GB hard drives, yet hardly any PDAs come with this type of storage. I also wasn’t keen on installing one of the Linux distros on a device that wasn’t necessarily designed by the manufacturer to support it. Even though I’ve got plenty of experience doing that on regular PCs, the specific requirements for a special, small PDA-like device felt like something I didn’t want to spend time trying to understand.

These requirements limited my search quite a bit. In fact, so many Linux-based PDA manufacturers seem to have disappeared over the years that I think there were really only three or four true contenders. I considered the following

Sharp’s Zaurus SL-C3200 (obviously) because it has a nice screen (though it wastes some space at the edges), decent hard drive, and space for memory cards and wireless/wired networking or other things. It runs Linux, (QT-based desktop interface), and seems to have a decent amount of community supporting its software.

Nokia’s 770, nice but it couldn’t fulfill my storage requirements and I read mixed reviews on its interface. Besides I’m used to KDE, so I assumed that a QT-based environment would be more my style.

The Pepper Pad, this seems like an exciting product but I think it should only cost about $600. I was suspicious about how much I’d like its style of keyboard and finally, as a relatively newer company, I had a hard time feeling comfortable about its future. Some of the non-conventional things they’re doing could have great results but if they disappear in a year or two I’d be left with something that I might be able to get much support for. In other words, maybe I didn’t look hard enough but I didn’t see much of a community around this product. Again, I might have been more willing to take a chance on this one had the price been more favourable to that kind of experimentation.

Archos PMA430–interesting, a media device plus. This seems like something designed as a media player but with a lot of the features of the Nokia product and the Sharp product. Unfortunately most of the reviews I read of this, made me leary about its functionality as the ultra-small computer/PDA I was seeking.

OQO 01+, this product just barely qualifies for my list (it’s more of a dream really). It’s the sort of device to elicit gross quantities of drool. Granted it’s the only one in this list that doesn’t come with Linux preinstalled (rather Windows) but it appears like it might welcome a reformat to Linux. I couldn’t realistically keep it on my list because it blasted through the high-end of my budget range. My impression is that the Zaurus is essentially a less powerful attempt at what the OQO provides at about a third the price.
So I was very happy with the little guy at first. Sharp did a mostly good job packing the functionality of a laptop computer into an ultra-small PDA style device. Its laptop form-to-PDA conversion form is a neat trick too (the screen swivels and auto-adjusts itself). The few gripes I had were that the wireless network card juts out into the space where you need to keep your hand in order to “thumb-type” and there was plenty of space to make the keys larger. The screen is quite crisp though it has unused edges that could have enlarged it. The battery seems able to last forever.

Installing software was not at all difficult (once you find some web sites that offer packages to install). Anyone familiar with Debian installation methods could easily handle the Zaurus. I don’t know why its office suite, Hancom Office, only supports Microsoft formats, it should also support OpenDocument Formats (ODF). Its built-in audio player was quite nice and useable but should have inherently support ogg formats in addition to MP3 too. Well this is my general complaint to all the software it comes with, lack of automatically supported formats. And all of this could have been easily solved, had the device been pre-setup with on-line package repositories, similar to Ubuntu or Linspire’s offerings. In fact, it seems like its installation app has all the fixings needed to do just that, but I couldn’t locate any such repository on-line.

I was able to work around the above criticisms, I don’t think they should be seen as condeming in any way, most are easily solved or are not really huge shortcomings so much as nice-to-have things. However, I found that perhaps this sort of device simply wasn’t the right choice for my needs. I essentially wanted a laptop computer that was easy to carry around anywhere. Even though laptops are supposed to be portable, the one I have is big and heavy and I hate carrying lots of baggage.

On the other hand, I don’t like carrying small devices for specific functions. I don’t really want a separate little PDA for contact and calendar information. I don’t want a separate audio player. Though the screen is sharp, I don’t like trying to read on such a small viewable area. I type fast, so the keyboard, although better than trying to fiddle with handwriting recognition, felt too slow for me. Thus the C3200, although it accommodates so much, ultimately wasn’t the right choice for me. I suppose a better choice for my needs would just be a cheap but very (perhaps screen size around 13″) laptop. And this leaves me to wonder how the market will treat the ultra-portable origami-style devices that have been making the rounds in the recent past.
——

This will be my last post here for a week. I’m heading off on a road trip from Montreal, to Fredericton, to Shediac, to Halifax, to Cape Breton, and beyond…

Bias and Time

By  | 

This is about consulting-in-the-world. :-) (excuse my weak, philosophy in-joke)

Paul Murphy posted a thought-provoking piece concerning consulting bias (plus), called Corporate loyalties and the temporal disconnect. He calls attention to the idea that people cannot really claim to be unbiased. We are wise to disclose bias so that we know how to deal with it and how it affects the decisions we make. I understood Murphy’s point to address more of the way people compare, for example, a product now with their experience of it in the past not taking into account the context of the comparison being “now.”

“…the memories haven’t changed, but circumstances have – and basing actions on comparisons in which one side is frozen in time is therefore intellectually dishonest.”

He elucidates this conclusion with several situations in which types of temporal bias would affect a decision. The post mostly is being asserted as the viewpoint of a consultant and clearly it is supposed to focus on a particular type of bias–the temporal sort, but of course there are other types of bias to take into consideration. That’s why I liked his point that

“The whole bias issue generally represents a fundamental mis-understanding of the problem evaluation process: open bias is often a positive thing…”

So, in thinking about these points, I had to reflect on TEC (the company for which I work). Our site frequently proclaims that we’re “impartial” and we attempt to present analysis of software data without bias. Is that really possible? On the one hand, I’d like to say yes because the way we evaluate the functionality of, say, an open source ERP vendor against a proprietary one is based (this is the most rudimentary way of saying it, actually it’s more involved) on a program that calculates features supported against those not fully supported. In other words this should take out the human bias that might be present in a consultant trying to recommend a system to its client. The consultant may be susceptible to the temporal situation pointed out by Mr. Murphy or, more likely, might be involved in a certain business relationship with vendors that provides an incentive for recommending those vendors’ solutions. Our company on the other hand, has no alliance to any particular vendor.

However, I have to think that using a program to weigh functional capability and a lack of alignment with specific vendors, do not necessarily equate to a lack of bias. At some level we could probably discover some form of bias. For example, as our analysts model the criteria on which to evaluate vendors because they rely on their research and experience they probably introduce, however innocently, certain biases. I can think of one simple example right away. Before open source software became a well-known enterprise commodity, many of our analyses did not include criteria for open source database support, thus in some ways, perhaps proprietary solutions had a form of advantage.

I would hesitate to call this criticism but rather I think this may highlight a reason why analysts, consultants, etc. have to be constantly self-critical, constantly trying to reflect on why they conclude certain criteria are applicable toward software comparisons and merit further research or recommendation. Thinking on the processes we undertake to form these analyses, comparisons, or conclusions may also be enlightening toward trends of the times. And that comes back to Murphy’s point on the changing circumstances of temporality.

Company Acquisition is Customer Acquisition

By  | 

A lucid read from AMR Research (I found this by way of The ERP Graveyard), which nicely discusses issues involved in enterprise software consolidation. I’m linking to this because I just mentioned in my previous post, the notion that the Infors (Golden Gate Capitals) of the world may be buying all the enterprise software vendors they can in order to accumulate maintenance customers and thus the revenue from those customers.

In some ways this may help the customers and AMR lists the reasons but there are also many problems this presents for customers. According to the AMR article, the strategy is likely concerned with packaging large customer bases (for future sale) rather than rounding out product functionality.

…this type of consolidation comes at a price that, unfortunately, customers must pay. By acquiring many competing products, aggregators tend to reduce competition within a market and cast the pall of acquisition over the remaining players. Because it is very expensive for customers to switch their enterprise applications, most companies have little choice but to stay put, pay maintenance, and hope for the best. While they may have had great leverage with the small vendor that originally sold them their software, they have little with their aggregator today.

So if the majority of a company’s IT budget is going to old projects I suppose this could be quite a concern. Unfortunately, and this may have been beyond the scope of the article, AMR did not really comment on how this affects innovation in the products.

More Noticings from LinuxWorld Expo SF

By  | 

After attending LinuxWorld Expos for a few years, I noticed a certain trend starting… that is, desktop Linux usage appears to be visibly on the rise. Considering I was attending a Linux-centric show, one would think that’s a given–but it’s not. I’ve been paying attention, informally (this is no scientific survey), to what people use at these shows. At many of the previous LinuxWorld Expos and Open Source Business Conferences I saw that not only a lot, but probably the majority of the attendees were using Windows-based laptops and more interesting was that a significant number of the presenters were using Microsoft Powerpoint on Windows for their slides.

With the variety of companies present at these shows not all of them should be expected to be pure-Linux or FOSS shops. A lot are providing a particular sort of software that happens to support Linux among other OSes, so it’s not entirely surprising that they’d be running OSes other than Linux. Still, at a Linux show, they ought to put their best feet forward. Anyhow, this time there were quite a few presenters showing slides with OpenOffice running on Linux. As I spied attendees typing away on their laptops, I saw that a lot were also running Linux. The predominant distro appeared to be a version of SUSE. Does the SUSE dominance portend something in the future of Linux-based business desktops?

As an aside, Linspire tried pumping the exhibition center full of its pheromones and jumped everyone in site, thrusting shiny boxed copies of its distro (with free manuals and other goodies). Always curious, I accepted. Linspire has no aspirations to the enterprise market, focusing solely on the home user, which makes them a bit of an anomaly among major Linux distros. Even other easy desktop distros like Ubuntu, Mandriva, and Xandros seem to have a few enterprise aims.

The Creative Commons had some great promotional items (and a well-done bit of DVD propaganda). My donation got me a snappy shirt that would either please Campbell’s soup or Andy Warhol–it’s hard to say which would have been more impressed. I was also instructed to take not one pin, but whole handfuls of pins until blood gushed from my palms. The trick now will be distributing them, maybe I’ll place them on the seats of some Metro cars.

And what about the enterprise apps? It would be nice to see more variety in the way of FOSS enterprise applications at LinuxWorld Expos (IT management, security, and development-related apps feel dominant, although as Linux gains home users, I wonder how the show will change to cover their interests?) A few of the enterprise app high-points included OpenBravo, which is an interesting new open source ERP system. I’ll have more on them (via TEC) in a bit. CentricCRM was visible and trying to show how it could be more appropriate for a large enterprise than its popular open source sister, SugarCRM. Finally, I’d like to mention Adaptive Planning, which although it has been in business for some time using an ASP model, just released its business performance management software under an open source license several days before the show. Adaptive Planning now is offering an on-line service as well as an on-site implementation with several different commercial options.

As a last note, in attending a presentation by Ingres’s CTO and strategy VP, Dave Dargo, he made a point about something like up to 80% of most IT organizations’ budgets go toward maintenance. If I understood him correctly, I think he was making a point about how the many existing proprietary vendors aren’t really interested much in new innovations with their customers in-mind, rather there is a lot of incentive to focus on all the clients spending money on old projects. I reflected on some of the recent acquisition press, and maybe I’m slow to have this really sink in, maybe it’s been quite obvious, but if this is the case then it puts a certain perspective on the motivation behind the Oracles, the Infors, etc. gobbling up so many competing companies. Does it matter so much whether they can offer a better solution for the customers of these gobbled-up companies? If they can integrate the wide range of systems? Is it mostly just important to them to milk a larger stable of maintenance customers?

A Few Noticings at LinuxWorld SF

By  | 

I had a few small problems getting in. They don’t care much for analysts masquerading as wizened party crashers with large yellow balloons depicting the faces of thousands of homeless linux users who’d like to make Moscone Center their home.

Not that I’d do that.

Virtualization seems to be quite the theme these days. The last LinuxWorld I went to was in Toronto and it was a very different experience. The Toronto show seemed a bit lackluster in comparison… however it felt more personal and had one of the most interesting keynotes I’ve ever attended. Namely the fascinating explanation of National Geographic and IBM’s genographic mapping project (can’t wait to participate). Of course that was rather loosely related. SF LinuxWorld and virtualization… After listening to several talks I don’t think I see quite what makes this such a hot topic… some cost savings in CPU licenses, etc. what else though?

There was a very interesting panel discussion on legal FUD with Eben Moglen, Christine Martino, and Stuart Cohen. It was a good combination especially when it came to the GPL v3 issues. The commentary was more multi-faceted and less extreme sounding then it’s sometimes portrayed in popular media articles. In particular it wasn’t as though HP came up heavily in contradiction to the draft content… much more in the spirit of recognizing the collaborative work-in-progress nature. In addition none of the parties seemed to feel this draft process was causing any problems with the development community–noting FOSS dev was continuing just as rapidly as ever. Moglen noted that it was all going according to schedule and he expected things to stay on course for completion in 2007 as originally intended. Why has there been so much media fuss acting as though the drafting is moving too slowly? Maybe that signifies a lack of background research and too many people repeating one another.

Another interesting bit were the database situations… EnterpriseDB winning best DB (nice for them and PostgreSQL) but also the presence of Ingres. Ingres now indepedant (mostly) from CA is an open source database to watch. They have thousands of customers and a long solid history to back them but at the moment they’re pretty much a CA-only solution. So it will be interesting to see how their new-found visibility and open source licensing will help them expand beyond CA’s shadow and get supported by some of the other major enterprise solutions. When I talk to open source enterprise vendors I’m still only hearing support for the major proprietary datbases and PostgreSQL or MySQL.

Finally, before I wrap this up, I’ll note that this is my first post from my new Sharp Zaurus c3200, which arrived just before the show. It’s a great little Linux device! I’ll say more about it later and I’ll have another post on LinuxWorld SF noticings.

What’s Going on with SMB Linux Accounting?

By  | 

WebCPA published an overview of some of the issues involved in Linux and open source deployments for the SMB crowd. It mostly focuses on some of the financial packages available, mentioning companies like ACCPAC/Sage, Open Systems, and InsynQ as offering their solutions for a Linux platform. The article, while providing what I thought was a pretty wide-ranging overview, seemed blurry on a few points though (noting, for example, that there would soon be Macs able to run Linux natively–actually that’s been possible for a long time, and mentioned freeware when I believe it probably meant free software).

The main thing that interested me was its discussion the companies that provide accounting solutions for open source platforms, most of those recognized in the article, actually seem to fall a bit short. Indeed, to quote a comment from an ACCPAC reseller presented in the article

“…Beck says most Linux installations involve running IBM DB2 database on a Linux server while running the accounting application on a Microsoft system.”

I wonder why that is? Shortly after, the article states the following

“Sage Accpac, which supports both Windows and Linux, estimates this year that as much as 20 percent of its installations were on Linux, up from 12 percent a year earlier.”

If that’s the case, I wonder why they’re not putting the complete system on Linux as opposed to setting up a mixed environment? Some of the arguments presented say that “the midmarket is not asking for it.” Is it that straightforward? I wonder if that means they’re only considering people that specifically say they want Linux. What if people want benefits, cost, support levels, functionality etc. that may be provided by a Linux platform but have not thought specifically to demand that via Linux? I have the impression that answering “the midmarket is not asking for it” seems a little incongruous with the rest of the stats about its deployment. I’d like to know how these match or do not match.

Finally, although the article highlights some proprietary financial apps that run with Linux or MySQL, it doesn’t mention a single financial application that is, itself open source. They exist though, perhaps I’ll follow up on this in the future.

GPLv3 and Corporate Contrarian Hype

By  | 

The latest draft of the third GPL version is provoking a lot of argument, posturing, and controversy. I’m glad its careful drafting process is taking the amount of time it is. I think it’s useful to widen the sphere of public awareness on the issues the license addresses. Some of the most controversial issues, such as digital rights/restriction management (DRM) and patents are going to impact our lives and culture in far reaching ways (they’re not isolated from technical and business issues). Yet a lot of people discussing these issues don’t seem to apply the rigourous thinking that is required.

I’d argue the FSF has a track record of considering important issues like these, with foresight and the creative will to develop strong, practical solutions, staving off potential damage to our freedoms. Damage that would otherwise be carried out by imaginary legal entities armed with human bullets, which fly toward profit so quickly they miss all other practical and ethical issues. The solutions have also enabled a huge amount of innovation and positive change.

But some of the most visible contrarians to this draft of the next GPL–opinions which are getting hyped, are mostly irrelevant in the greater scheme of things. I’m talking about the recent HP issues that were circulated around numerous web sites. To quote Christine Martino, vice president of Open Source And Linux with HP from the Internetnews.com article linked above,

“HP had hoped that the second draft would clarify the patent provision such as to ease concern that mere distribution of a single copy of GPL-licensed software might have significant adverse IP impact on a company…”

What does that mean, “significant adverse IP impact”? It’s removed from its context so I can’t be sure, but it sounds to me like the HP folks are taking issue with something the new GPL would prevent them from doing with their patent portfolio. Furthermore, referring to HP’s commentary, the Internetnews.com article states

“The second draft of the GPL version 3 license is not even a day old and already one of the largest Linux vendors in the world is taking issue with its content.”

So what? The FSF is interested in freedom, and its foresight in ensuring and encouraging that was the ultimate basis of practicality giving rise to the IT business shifts underway because of FOSS. Although the Open Source Initiative fairly claims the pragmatic approach under its rubric (as that is its stated goal), it doesn’t imply an either/or stance. Freedom doesn’t preclude pragmatism. Unfortunately too many articles treat these notions as mutually exclusive.

While the previous GPL versions were produced with a goal of promoting freedom in a software development basis, they also triggered important business and social developments. Why were they so successful? Because many many many individuals adopted these licenses. The freedom and collaboration they enabled for masses of individuals in free software development is key. A recent ZDNet blog post states

“But that’s not where the debate will really play out. It will really play out in the market. Will GPL companies switch to GPL v3, or explicitly demand retention of V2, which is frankly vague on the DRM question.”

Does it matter if major business entities like HP object to certain freedom promoting aspects of the license? Does their wanting to switch really signify whether the third version will be successful in getting adopted? I don’t think it matters much at all. HP is a player in the free software community, it is not the player. And that is true of everyone else. So HP commentary should be considered for what it’s worth–nothing more.

So, I’d argue against Steven J. Vaughan-Nichols’s point that the GPLv3 will be dead on arrival. As he mentions the HP issue, he also mentions Linus Torvalds’s objections. This is fair, from what I’ve read it sounds like Torvalds has some clearly thought-out opinions. From what I’ve read, some of these sound quite reasonable. As I said at the beginning, I like the debate this drafting process is raising, Torvalds and HP included. Nevertheless, from what I’ve read of Torvalds’s arguments, I have the impression he is single-mindedly focusing on technical issues and intentionally excluding debate on all else. I just don’t believe that’s ok. There are too many important, non-technical ramifications interconnected with information technology to ignore. It doesn’t mean everyone must think about these things, but how does it help to intentionally excise them from the debate?

——
Addendum

A new story covering the feedback issue in this debate was just published at NewsForge. This is a useful balance to the different sides involved.