Small Server Outage

Small outage of this server yesterday.  Just one of those mysterious crashes, and then the networking fails to come back up.  Very frustrating and mysterious, because, as far as I can see, all the correct scripts are being run.

Not going to be the highest priority in my life, because I am planning to move the blogs and wikis to another hosting setup within the next few weeks, anyway, in an attempt to reduce costs.  Right no my main focus is on the garden and getting enough garden beds prepared for the Summer.

Vista Sound

Zoli has an entertaining little anecdote over on his blog about how users cannot disable the startup sound in the current (beta) release of Microsoft's Vista OS.

Funny, back in the days when I still had a Windows machine (around 5 years ago), I never could be bothered to figure my way through their twisty-little-maze of configuration dialogues-for-the-brain-dead to switch off the startup noise.  I used to just jump in and delete the damn media file.

So I guess there is a way to turn it off after all...

Social Networking 2.0?

A thing that really, really irritates me about the whole "social networking" hoopla is how it takes such a short view of history.  Dave Pollard has a very nice, very useful writeup about social networking (2.0-style).  (Sidebar: Was that mindmap done using Freemind, Dave?  Great piece of software!)

But let's not forget that we humans have been "social networking" since before we fell out the trees.  The fact that we're now trying to do so over the 'net, through a much-lower-bandwidth interface than we're geared for (nothing beats face-to-face!) just means that we've set ourselves some obstacles to overcome.

Perhaps the only significant contribution that the 'net brings is the ability to communicate anonymously.  Some might argue that this is also the handicap that the 'net brings to our conversations, too, since it is what enables spam, wiki defacements, etc.  But I think that anonymity is also exactly what allows us to express a lot more of our true nature, our inner self.

How to Screw-Up Your Web2 Application

If I were a marketing guy, I would keep you in suspense right up to the end of this post.  I would waffle on for ages about how and why I'm going to tell you "the secret," and what a super guy I am for letting you in on this.

But I'm a programmer, and time is precious.  All over the web I see this particular piece of egregious stupidity:

Apps that use email addresses as user-ids.

I strongly advise against the use of email addresses as login ids.  Consider the following 2 common cases:

1. The user changes their email address (due to changing provider or whatever). 
2. A user leaves the community.  Months/years later another user joins; they have the same email address as the old user, but are not the same person.  Are you going to refuse them entry?

In the former case, if you allow them to change their login to correspond to their new email address, you lose the trail of what they've done over time, since you've essentially changed their identity.

Worse, yet, if you're doing any kind of app that allows the user to build up a history, karma points, reputation, whatever, since you force them to throw away their entire investment in your site.  They may as well go elsewhere.  That history took the user time, energy and effort to build, and constitutes your only real barrier to entry against competitors who want to eat your userbase.

In Summary:

A login-id is an identity.  An email address is not an identity.  It is an address.

Microsoft playing catchup to Jini?

I just saw The Ray Ozzie Experience and almost ROTFL.

"A world of many devices, all connected and managed by the Web".

Isn't this what Scott McNealy was telling us like eight years ago? Remember "WebTone"? Welcome to the party, Ray Ozzie.  Sorry that you're so late! M'afraid the beer's justabout finished...

Isn't this exactly what Jini was designed for? (And I'd infinitely sooner bet my life on technology as mature and carefully designed as Jini than on anything MS is ever likely to come up with.)

Why is CSS so damn HARD?

Seems to me that the whole CSS model is pretty poorly designed.  It shouldn't be so damn hard to implement a website design.  I'm not talking about bleeding-edge Zen Garden stuff; I'm talking about very simple layouts.

For a start I prefer liquid layouts: That graphic designers coming from more traditional media hate fear and loath the concept, I understand.  Its a mindset - the user has partial control over how a thing looks - and many graphic designers have trouble dealing with their inability to guarantee pixel-perfect alignments.  Perhaps the user wears hectic prescription glasses, so 18pt fonts are a reasonable default for them.  Get over it.

Secondly, I'm no n00b at CSS.  Whilst I'm hardly a professional CSS designer, I think I understand the concepts and details pretty well, and I've fumbled my way around a fair number of web designs using CSS with results that have attracted fair compliment from people who do that stuff professionally. (No, this blog is not currently an example! That's what I'm working on.)

But its still so damn hard!

One of two thing I think are needed: either

1. a redesign of CSS that works to a "springs 'n' struts" layout model, or alternatively a "springs 'n' struts" model that can get compiled to CSS2 (possibly on the fly as a filter), or
2. the additional of another "position" mode in CSS - "absolute-relative" positioning - absolute positioning of an element, but relative to the containing box.

Number one is unlikely (except maybe as a translated/compiled language), but number 2 is possible without breaking existing CSS-based layouts.

It would sure make simple layouts a hell of a lot simpler to implement.

Reports of the Death of Email Greatly Exagerated

So there have been a couple of surveys among college students indicating that the surveyed population mostly uses IM to keep in touch with their social circle.  They only ever use email to contact companies and "old people" like their parents, and view email as "old fashioned".  This has led some commentators to pronounce the Imminent Demise of Email as communication channel in the Internet.

Horseshit!

Social Networks are Killing Email? - I think not! What we are seeing here is the confluence of two things:

1. Younger people have more time on their hands, and so are more inclined to spend some or much of that time in online communities, or social networks. (Newsflash: Social networks are nothing new.  We've been doing social networks since before we fell out the trees!)  As a result they tend not to use email, preferring IM, as email lacks an immediacy - we're seeing the impatience of youth.
2. As one ages, one's priorities, as well as the demands on one's time, change.  Therefore, the preferred modes and channels of communication one favours are likely to change with age.  Duh.  Why do I love my cellphone more than my landline?  It does voicemail.  I can disconnect without missing anything.  There are times - many times - when I prefer to disconnect.  I truly won't miss anything important in the endless torrent of attention-grabbing shit, and I need time to myself.  Time to stop and think. Time to reflect.  And I'm not insecure about it.

So, in a nutshell, we see 1) the impatience of being young, and 2) the insecurity of being young, reflected in some surveys that say that young people don't much use email.

And from this, people are extrapolating Email is Dead?  Let's see the same surveys done on the same populations 5, 10, 20 years from now, and lets see how their channels of communication have changed then.  Perhaps then we can begin to draw some conclusions, instead of this bogus pseudo-science.

Otherwise I will stick to my premise: Death of Email Greatly Exagerated.

Disclaimer: I am in my mid-40's and have two sons who fall into the "youth" category, one at Rhodes University, the other a hotshot Software Developer.  As such, my views may be biased.