Tuesday 22 February 2011

@Twitter Vs @Twitdroyd

So this weekend was an interesting one in the twittersphere. On Friday twitter decided to suspend UberMedia client products for some rather unclear reasons.

Now for years there has been tension between platform owners and platform access gateways. With the company behind @twidroyd rumoured to 'gate' up to 20% of twitter usage there are clearly bigger games at play.

So what did I personally learn from & get reminded about from this :-

1) my main usage of twitter is now via my Android Nexus One phone

2) Twidroyd is by far the best android twitter application - I tried twitter www mobile, twitter native app, tweetdeck & tweetcaster. Whilst tweetcast was the best it was still frustratingly poor compared to @twidroyd

3) I now use twitter as my first point of call to gauge the pulse on both the IT industry & world events

4) That I spend money, and am now 'dependent' on, cloud services that I have no SLA or control over

A minor blip for sure, but I'm expecting quite a few more of these in the future as providers go through the wrangling re who "owns" the customer & the data they create...

Thursday 17 February 2011

End of availability

Just a quick post this morning, no it's not about my seemingly new role as "GrumpyHermit" but rather about vendor product churn.

You see in the last two weeks I've had two supposedly global 'top 5' IT infrastructure vendors casually inform us that four of the products we standardise on worldwide will no longer be purchasable in ten days time!

Now naturally we have agreements in place re minimum 18 months availability, and 6 months advanced notice of any EOA date - but nowadays these get treated with the corporate equivalent of "meh" by suppliers...

Naturally this has all sorts of impacts to WIP projects, our design & build teams and our competency centres re build images & operational readiness acceptance etc.

This is further compounded by the vendors increasingly operating "build to order" policies with little or no stock held in channels, thus causing inevitable delays around each sku change. Of course it doesn't help that this increased rate of change appears to be hand in hand with a reduction in core quality & testing by suppliers.

So please vendors I implore you to remember that diversity is the enemy of efficiency and drives real world customer costs through the roof... Please give some serious thought as to how to abstract the customer from the now constant hardware & firmware changes masked in so called 'benefits' (it's 5% faster, 7% greener, 3% cheaper etc) that need new configs, new drivers, new testing, different interop skus etc.

From my perspective, if its not a "direct field replacement unit compatible" then it generates real cost and pain - not dissimilar to that of changing suppliers...

Oh and whilst I'm on the topic, why isn't there a standard (ISO, DTMF, ANSI etc) on product lifecycle definitions & lifecycle events?? And don't even get me started on the concept of major & minor product version numbering standards and conventions between products & vendors...

The current situation on all these areas is chaotic, unacceptable and untenable. The IT industry is slowly killing itself through unwanted & unwarranted 'change for changes sake' that is driving a real but often unreported cost time bomb...