19 January 2006

Betfair, Australia… and technology?

Everyone who follows it is up to speed on betfair and their recent licensing in Australia. However, Betfair corporate and business affairs director Andrew Twaits has been paraphrased as saying that “the core function of the business was technology and the Technopark centre would not be a call centre, which was a common misconception.”

Next, factor in that the Economic Development Minister, Lara Giddings, was paraphrased as saying that the new Hobart betfair opreation “was expected to have 150 jobs at the end of two years and to employ 75 by the end of the first 12 months”.

Doesn’t that now beg the question as to what 75-150 technologists are going to be doing? Betfair already has at least 150 technologists (perhaps 1/3 of staff) working on their staff in London. I would have guessed that any regional offices/BU that betfair opened up would have been primarily a call and customer support center, with some bizdev and regional marketing thrown in.

A few possibilities:

  • Betfair is making misleading statements so that Australians can’t complain about the type of jobs that are being created (high tech, high paying technology versus lower paying call center)
  • Betfair is shifting some portion of their technology from London to Australia. Australia, particularly Hobart, offers significantly cheaper salaries and costs than London
  • The technology build up in Australia is purely a de-risking duplication and additional build-up of technology capability.
  • Australia provides a foothold into SE Asia - the technology team will be focused on building up Asia-focused versions of the product
  • Betfair made related commitments to the Tazmanian government, and is now verbally following through with them. Over the next year or so the commitment will wane unless one or more of the above is correct. Betfair will claim “changing market conditions” to pull back from their commitment.

16 January 2006

Browsing with flock and using web services

The most useful technology I've come across recently is flock. Flock is a derivative of the Mozilla/Firefox browser. Flock's feature twist is that it allows greater integration into a growing number of online collaborative services (an aspect of the so called Web 2.0). It's not so much flock's greatness (flock is certainly good anyway), its that it has opened the door for me to new online services and simplified and encouraged their use. I believe that killer features (like the ones I list below) will become commodities and more progressive browsers like Firefox (and last generation browsers like MS IE) will fold the features in. The key is the seamless integration and use of the new features, regardless of how you get to the features.

So what are "online collaborative services"? Briefly, they are an online information sharing service that you would normally access via a browser but can also access via an Applications Programming Interface (API). It is the API access that allows applications like flock to tightly integrate to the service. And by tight integration, I mean that you can use the service "in context", that is, you don't have to "go anyplace" (e.g., another webpage, login, navigate, ...) to perform an action. You just have to do the bare minimum to complete a task. To understand this better, consider the following online services and flock features.

Service/Feature #1 - del.icio.us and favorites

del.icio.us is a service that you use to specify and store a set of links and share them with others. The community that adds and shares links results in popularity and discovery of shared interest information sources within the community. While handly for users that frequently move between machines they don't have control over (e.g., Internet Cafe, University), I didn't consider it useful as I move between work, home and laptop, all of which I have sole control over. While a common links page was mildly useful, it wasn't useful enough to flip to a dedicated website to do this.

Flock directly integrates its "Favorates" function into del.icio.us. Therefore, no matter which browser you use flock from, your Favorites list is synchronized at flock start-up and whenever you add a new favorite. This solves the multiple favorites list when switching between browsers at work, home, and laptop devices.

Flock also provides enhanced search functionality, which includes a search of your favorites before using something like google (more on this below).

Privacy should be a concern that comes immediately to mind. del.icio.us generally shares your favorites with the del.icio.us community (partial work-around exists with explanation). If someone knows your uid, they access your list of links and tags (keywords you use to identify page content when you add a new link into del.icio.us). On the one hand, there are good reasons for congruency throughout ones life (private-public, work-home). On the other hand, business competition is a reality, and I wouldn't create a publicly viewable set of tagged (grouped) links for a sector for which I wanted to maintain secrecy and competitive advantage. Not to mention pr0n preferences!

Service/Feature #2 - blogging

Blogging is the act of writing about and sharing information. Often blogging includes references to another source of information.

Flock includes direct integration to three blogging services - wordpress, blogger, and typepad. Having never blogged before, I started with Wordpress, then moved on to blogger (see my previous entry for rationale). After setting up an account, feeding the uid/pw to flock, I was able to make immediate in-context blog entries.

I'll stay well away from judgment or thoughts on blogging, at least for now.

Service/Feature #3 - feed reader

Many websites, especially news portals and blogs, provide information feeds. These feeds provide a flow of new information being posted on the website. Items in the flow may only contain summary information, or may be the whole chunk of information as it is posted on the website. A feed reader is an application that allows you to specify which feeds you're interested in, then present those feeds to you. The format and manipulation of the data varies by feed reader application.

Flock (as inhereted from Firefox) provides a feed reader, and is clever enough to look for feed presence indicators on each page it loads. If it finds such an indicator, it notes it in the address bar with a Feed icon. Also, if you mark a page a Favoriate that has a related feed, flock will automatically start downloading and saving that feed local to your browser so that you can view it anytime (connected to the Internet or not). The feed reader is pretty unsophisticated, but if you're not using one yet, flock provides a soft and easy introduction.

Service/Feature #4 - flickr

flickr, purchased not too long ago by yahoo, is a website that allows you to upload and share photographs. There are many other similar websites, with the first one I ever used being ofoto (now Kodak Gallary). Other than the API flickr offers (which enables flock integration), I'm not sure why flickr is seeing recent popularity.

Flock claims to provide good integration to flickr.

While I've downloaded and played with flickr, and flock-flickr integration, I've hit four roadblocks with it:

- flickr (as a free service) crippled your monthly upload compared to other similar services

- I don't care too much for the interface

- The flock-flickr integration doesn't work, at least for me (flock never displays the pix I've uploaded to flickr as part of trying it)

- I'm not a big photo enthusiast (I take lots of photos, I just don't do much with them!)

Feature #5 - search

This isn't a web service, but its still a very useful feature. Like Firefox, flock provides a search text entry box. However, Flock cleverly searches your history and favorites as you type in text, and displays those results in context (dropdown of sorts below search box). Now that I've built up a link history within flock and a Favorites set within del.icio.us/flock, I use this feature constantly.

Feature #6 - the shelf

The shelf is like a cross between an in-your-face set of favorites and an application neutral clipboard (where your copy/cut-and-paste content sits). The shelf allows you to highlight-copy-paste items that you find while browsing that you want to do something with later. The shelf sits as independent window, and accumlates (like a clipboard) items you drag or copy into it. I've not used this feature much, although I could see if you approach blogs with the approach to "gather-file-gather-file, write about it, bin files", it would work fine.


How solid is flock? I've not had it crash once. I have seen it start to freeze up between opening new links in new tabs, and seen memory use climb to silly levels (560MB in use). The blog entry editor is a little buggy, so you might have to do some manual editing within the blogging site interface to fix your post. There are other niggly details like error reporting and assistance/diagnostics, but nothing big. If you don't mind quiting and restarting periodically (perhaps daily or weekly depending on your use level), I've not had any significant problem.

What if flock and its development team goes away? Well, while you might get spoiled by the ease of direct service integration and other nice features, your investment isn't lost in the services you use. All of those services are available via their websites, and you can always use them that way. Other aspects, like all of your links that sit in del.icio.us, are cached to flock so you always have all your Favorites data anyway.

Realistically, I don't see the commercial relevance of flock. The more useful features will either be merged into Firefox, or maybe even eventually MS-IE. However, open source communities/products like Firefox, and tangential efforts like flock push the innovation envelop hard. I look forward to clever new web services and tight integration with the browser (and someday desktop). I do appreciate flock team efforts, and I hope they have any number of clever feature and commercial tricks up their sleeve to make a go of their business.

I've not been using flock and its new-to-me features for long now, so I don't claim great authority on it. This blog entry is a YMMV analysis to date using flock and the services it connects to. Regardless, I'm glad I found flock and spent time learning about the services it touches. I have found flock sufficiently stable and compelling enough to use now as my primary browser.

blogspot versus wordpress - a mini review

I am debating a switch of my (not so old) blog from wordpress to blogspot/blogger. The main reason is that after playing a bit with each, I feel I have more control over content and layout with blogspot.

I've only briefly examined and used both, so my debate thus far isn't highly considered or scientific. Summarizing my experience thus far:

- Use wordpress.com if you are not a power user and will happily sacrifice flexibility for simplicity. There are simple tools to change all the aspects of the page.

- Use blogspot.com / blogger.com if you are more of a power user. The tools provide greater flexibility but of course are less simple (e.g., can directly edit HTML to change your list of links).

A few other things I like about blogspot that weren't immediately obvious in wordpress:

- blogspot also gives me a simple interface to google adsense, which I wanted to play with

- I can email in blog posts and they automatically publish

A few things I don't like so far about blogspot/blogger:

- blogspot/blogger is that it is sometimes painfully slow compared to wordpress

- I would appreciate tools to manage stanard plugins like lists of links rather than having to directly edit HTML

- You have to manually publish a new blog entry rather than having it automatically publish like new entries in wordpress

I've not made up my mind yet, so I'll double-post here and there until I settle on one of the two.

Installing SuSE 10.0 Linux 2.6.13-15.7 on an IBM laptop X31 - GRUB fails, error 18

While installating SuSE 10.0 Linux on an IBM Laptop X31, after installation and then at first boot, I received a GRUB error 18.

My first effort was to reinstall SuSE Linux. Same problem.

While researching the GRUB error, I noted several entries that discussed a problem with older versions of a BIOS being used with HDDs that were too big (more than 1GB). Given that the hardware (all stock) and BIOS I was using were up-to-date, this didn’t seem overly relevant.

For kicks, I decided to try forcing the boot image into the first 1024MB of the HDD. As my SWAP was the first partition on the drive, and it was about 1.5GB (768MB RAM), I thought perhaps this might be causing the problem (my “/boot” area, as part of the default “/” partition, was well above the first 1024MB of the HDD).

Ah ha! While this wasn’t overly scientific, I can now say that after shuffling around the partitions, my IBM X31 laptop now boots with Linux.

Google pack and why

Since the Google “Pack” was announced, there has been a general criticism about its utility. Was it folly on Google’s part to make and release the pack? I think not.

Assume that Google is going toe to toe with Microsoft. As Google inexorably builds the core competencies required to do this, it must build up the ability to build and release software to the public. Google started with being a purely web-based offering with no heavy clients required. Clearly this product type only allowed Google to deliver relatively simple web-based applications. Then, with the addition of products like Google Earth, they started offering (internet enabled) client side applications. This enables Google to build up competencies in the production and distribution of heavy client applications. The next logical step is to then build up competencies in how to build up not just the distribution of one piece of software, but an integrated set of software products that provides an overall solution for a broad segment of customers.

Is the Google Pack innovative? No. Does it bundle together a useful and compelling set of software? No. Are any reviewers praising it? I’ve not heard any yet. Will it get many downloads? A few. Did it enable Google to softly build competencies in the delivery of a bundled set of software? Oh yes.

Poker robot mitigation

A quick and simple solution… If companies like Party Gaming/Poker were serious about cutting out automated/robot play, why not insert a captcha function (word verification required) every so many games?

Captcha is used by portals like Yahoo, email sites like Yahoo, and domain registrants like godaddy to block automated account creation. It is the phrase you read from a “warped” graphic image and type in during the registration sequence.

If automation/’bots are blocked by these mainstream sites, why wouldn’t it work for poker? What am I missing?