25 Jul 2009
(East) Africa just had their global Internet connectivity significantly expanded. Education applications are presently limited to the tertiary sector. However, the promise of growth in Kenya and Tanzania particularly is significant as costs fall. Initially ISPs in this region have gone for higher bandwidth over cost reduction. That said, if Internet access developments follow models established already in China and India, conventional ISPs aren’t going to deliver growth, mobile providers will.
Accordingly, the improved bandwidth situation at the present prohibitively expensive costs of ~$600/month for a good link is ultimately a bit irrelevant if mobile tech delivers last-mile infrastructure and the mobile web enables e-commerce, social media participation, governance, healthcare and more. This isn’t a case for existing ISPs to drop prices: they’ve definitely got a very good business case for leaving prices up but using the link to improve value while this is still a valuable commodity. The only significant short-term challenge to this comes, potentially, in the form of any government policy implemented. They might do well to intervene here and stimulate economic development by promoting global connectivity… but I suspect the interests of established business and government, if they resemble anything like those in Australia, coincide too significantly for such bold maneuvers to ever come to fruition!
From a business standpoint, it makes sense to capture these markets with medium bandwidth technologies early. That said, the relatively limited capacity of this additional global link makes co-location essential for any serious engagement. What this represents is an important in-road for low-outlay development of new markets with significant parallels to existing products (i.e. to English-speaking populations without need for additional infrastructure).
For East Africans, however, this is much bigger. Internet connectivity enables exports of innovative solutions, and, as social media uptake improves, of localised (l10n)/internationalised (i18n) solutions in response to this newly-visible Internet market segment. The problem of ghettoisation along language lines is not so prominent perhaps as a result of significant Anglophone influence — Francophone Africa will, of course, engage in different networks because of language barriers. Yet some services, Twitter perhaps eminent among them, have irrationally succeeded independently of ‘native’ language (it remains at present offered only in English and Japanese, despite significant Chinese membership, and, who can forget, Iranian political application!) — while others (Facebook, to pick a similar example) have languished and been replaced by clones despite their linguistic plurality (26 unique languages last I recall hearing a count, including English (Pirate) and many more serious ones) — Xiaonei being but one example of this.
If language is not an issue, it is possible other disparities will become divisive in the same way. Developmental barriers in terms of software industry (a key driver of domestic web innovation) and global trading partners will steer usage in any number of particular directions. For example, China’s inept attempts at achieving independence from Microsoft software in the last decade have been effectively squashed by their rampant piracy situation. Parts of eastern Africa engage in literal acts of piracy, but it’s probably not indicative of an attitude towards or developed industry against protection of intellectual property. If the criminal distribution network doesn’t yet exist, and software adoption is insufficiently mature, it’s entirely possible that open source could win. This is naive, and based on the presumption that Africa has, to date, existed in a vacuum — but if we consider for a moment a day working on a computer without Internet connectivity, something of the radical difference between minimal connectivity and full-on broadband enabled connectivity begins to sink in.
One Australian commentator recently observed, in response to a dramatic increase in average per-capita bandwidth consumption/annum, that there are a number of “tipping points” in Internet usage. For example, in the last 18 months, availability of online services as well as wider adoption of home broadband has resulted in a massive expansion of data transfers despite only a marginal increase in average connection speed. Youtube and its ilk have entered a perfect storm of gradually expanding connectivity: it just so happens that at certain points, connectivity results in usage peaks (which then plateau but don’t decline) as consumers discover new ways of using the Internet to interact. This happens with the transition from dialup to always-on Internet, and it happens again at certain speed points–consider tabbed browsing as well as video on demand/what we now consider “bandwidth intensive” activities.
This could be a tipping point for economic development and global integration. Watch closely!
21 Jul 2009
Spent a little while this afternoon cooking up a small website for SUEU‘s international student group, Focus. Born largely out of the realisation that http://www.sueu.org.au/activities/international_students/ is probably just a little too long to be catchy, it’s just a microsite designed to be relatively static (as the current page is) and redirect people to real-world activities.
Hopefully the new euFocus web address will serve well and be a tad more bite-sized/amenable to inclusion on posters, print media, etc. For usability, I’ve gone the non-geek route: http://www.eufocus.org/ (nofollow’d here) can pick up anyone who needs a normal-looking web addres via 302 redirect. Lovely.
Future ambitions may include online event rego and internationalisation fun times, as about 90% of Focus-ers are from mainland China. This will ideally take the form of user agent locale identification — it’s not mission critical and geolocation stuff is largely irrelevant for what is a very geographically constrained activity.
18 Jul 2009
I enjoyed the latest Harry Potter film last night — and it is the first in the series I am genuinely able to say that about. The rest have been frustrating for a variety of reasons. Some are technically poor, others are merely victims of terrible direction and script work.
The latest, however, has been remarked upon by various reviewers as taking a significantly darker turn — reflected both in (generally) well-done digital colour grading as well as thematically. The idea of “darkness” in this film is intriguing, particularly in light of the acknowledgement of Dumbledore as one of the greatest wizards of all time and the continuing struggle and dichotomy between forces of good and evil in the film.
The demise of Dumbledore (oh, yeah, spoiler alert — but you’ve already read the books, right?), that great force for good, represents a notable setback for this cause. I mention him particularly because he is, until this moment in the saga, the character with the most complete view of the situation. He is wisest and most powerful. However, as he enters the cave containing a horcrux with Harry, I was frustrated with their collective inability to illuminate the room.
Cry lumos as loudly as they may, even Dumbledore’s casting of beacons of light across the cave fails to significantly brighten the environment. I was reminded of the triumph over darkness that God has given us in Jesus, so absolute that fiction does not even approach its conquest. The Bible speaks of a future, renewed creation where God’s holy city has no need of a moon, or a sun, for it is so illuminated by the glory of His presence. (Revelation 21:23-24)
This is not fiction, and there is no battle that has not already been won by Jesus Christ the living king of all.
God has delivered us from the domain of darkness and transferred us to the kingdom of his beloved Son, in whom we have redemption, the forgiveness of sins.
– Colossians 1:13-14
…that you may proclaim the excellencies of Him who called you out of darkness into His marvelous light.
– 1 Peter 2:9
17 Jul 2009
Recently upgraded along with the rest of the world to WordPress 2.8.1 and hit a whole bunch of craziness that essentially encouraged me to move even this site across to WebFaction (eating your own dog food and all that) because of no longer being able to post from the WordPress admin panel.
I hadn’t changed any plugins, and the rest of the net was conspicuously quiet, so I figured that either my current host had near-simultaneously upgraded the database & corrupted something along the way (with MySQL this can result in read-only tables in certain cases, I’m told), or else there was something more sinister at work. No plugins had changed, and because nothing else functioned differently for even a moment I presumed innocence on their behalf. I tried with and without Google Gears enabled, with and without tags, in different saved statuses, and still nothing worked!
Anyway. Turns out it was my use of the term “wget”. (Even now I need to be careful as I write that!)
mod_security on Apache basically threw up its lunch every time I threatened to post about it. Accordingly, I can’t share with you some things that I’ve learnt in migration just yet, and haven’t been posting about anything else as I’ve been trying to resolve this problem, thinking it was preventing me from posting. Perhaps mod_security is trying to get me to write about more meaningful things than file transfer trickery! *yawn*
At any rate, it’s all over now and we’ll be back to regularly scheduled programming shortly.
10 Jul 2009
I’m currently in the process of trying to roll together a few hosting accounts of mine that have unnecessarily been running for the past few years as I’ve discovered ‘better’ services but not bothered rolling the old accounts over to. I’m probably losing around $700 a year because of this, and basically killing any revenue from hosting for other clients by making the whole process so unmanageable.
It’s frustrating, because it’s self-inflicted, relatively fixable, and a prime example of terrible stewardship on my part. It trickles out of my account in minor transactions of around $20, and I just don’t miss the money as much as I should. The problem is, if the transfer gets screwed up, various people dependent particularly on email (less so websites — they can withstand being down for a period of time, but for most clients of mine email is THE crucial application) cease to function until the DNS mends itself.
For this reason, I’ve not been brave (foolish?) enough to entrust the task to a service such as ODesk or RentACoder, even though either has the potential to totally take the headache away for a pretty minimal cost. I know that the odds of something going wrong between the exposure of core hosting passwords to strangers, in a process that is the web equivalent of a heart bypass, are pretty unacceptably high to simply palm off to some stranger for a chunk of change. Part of the problem is the kind of business continuity capabilities faced by many of my clients aren’t such that service notifications provide any particular benefit: it’s not as though they have any particular remedy for ‘scheduled maintenance’ in terms of notifying clients, as, unlike larger businesses, their websites are not frequently visited as first point of contact for vast numbers of existing and prospective clients of their own, and public apologies or notifications are meaningless.
Similarly, the scope of client education for such a minor undertaking is itself quite formidable — this sort of outage is highly occasional and the clients are so varied that there are no particular processes in place for dealing with it. Micro web agencies aren’t generally well equipped to do this sort of thing, simply as a byproduct of the nature of the provider/client relationship. In my work with larger businesses (especially where SaaS is a core offering) where the relationship is less provider/client and more embedded (i.e. I don’t end up functioning as an external party!) we have of course formulated plans for continuity and notification, but this cannot be the case as simply or readily for smaller, more fragmented organisations.
These issues have produced something of a perfect storm, where minor recurrent losses are the path of least resistance in a situation that requires a fair amount of (non-financial, tangible and intangible) investment to correct course. This, combined with the fact that I don’t have enough clients to justify writing migration code, and the general awfulness (particularly the glacial slowness)of WHM/other proprietary host management software, has meant I’ve yet to embark on an exercise with little visible benefit. Over 50% of this task is stuff you can’t outsource, or at least shouldn’t: client education and maintaining relationships.
Small business owners: how do you balance this need for process improvement and cost saving with the reality of day-to-day busyness and your obligations as a provider?