Re: [Linux-aus] Funding a real job database

[Anand Kumria <wildfire@progsoc.uts.edu.au>]

Hey Leon,

On Wed, Feb 19, 2003 at 06:19:34PM +0800, Leon Brooks wrote:
> On Wednesday 19 February 2003 04:49 pm, Anand Kumria wrote:
> > Hmm, so you are suggesting that LA change from a small group of people
> > into a 400-strong organisation with premises and an employee.
> 
> No premises, part-time employee.

More complexity for the treasurer. Yay!

> > Personally I think that is an extremely large transition to make -
> > especially since very basic matters; such as accepting new members are
> > yet to be addressed adequately.
> 
> If we don't do stuff in parallel we will still be fluffing around when they 
> plonk a space elevator down on the other side of Rottnest. (-:

If we had a really hot job market this might be interesting to do. We 
could probably even convince some hotshot Linux wannabe company that 
this is in their best interests to support us doing it.

However the best way to achieve "brand recognition" -- which is what I
think your fundamental goal is -- is still 'word of mouth'. Every other
suggestion (advertising on TV, Paper ads, etc.) does the same thing only
faster.

I think the question is 'Is speed of the essence?' for LA's jobs
database. If we don't act now, will it go away? The answer is that it
won't - so long as someone is willing to look after and maintain it.

That could be both a blessing and a curse! Anyway, despite the low 
frequency of job postings, at this time, it is still trucking along.

> >> And what should we offer employers (nominally for their money, but this
> >> applies even if we offer the service for free)? I'd suggest four levels
> >> of validity: none, scan-of-dox sighted, real dox sighted, and document
> >> background verified (e.g. by 'phone call to organisation originating
> >> document). This would allow the employer to select a risk level at which
> >> to operate, and it seems reasonable to me to charge employees/contractors
> >> a moderate amount of money to sight (and scan, and maybe follow up)
> >> identity and award documents.
> 
> > It isn't clear to me what documents you particularly have in mind that
> > would need to be sighted.
> 
> Sighting personal ID, paper certification (especially when they can't be 
> checked on line), contacting references and validating them to some extent. 
> Not all `employers' will need that level of detail, not all `employees' could 
> be bothered submitting it. But some of each will, if the opportunity is 
> there.

So you are suggesting that LA set itself up as some kind of trusted
third party in the job market; able to assuage employers concerns about
potential employees resumes?

So in reverse order of your points:

	- contacting references. I certainly don't submit my resumes
	  with my referees listed upon them. I'm not going to waste the
	  time of my referees until I know more about a potential
	  employer.

	  What value is a referee check when you go for a position
	  somewhere 12 months later. Very little. It'd mean constantly
	  refreshing that aspect of person's resume.

	  Time consuming for all parties involved with very little
	  benefit. To either potential employees, their referrees or
	  employers.

	- paper certification; as you note not all can be checked
	  online. But all of them require some kind of annual
	  (bi-annual) renewal anyway. I'd be highly surprised if those
	  that don't have an online facility at the moment aren't moving
	  to put on in place shortly.

	  For other paper certification, e.g. degrees, you've then got
	  to worry about issues of recognising an authentic University
	  seal. Worse, the effort involved is less valuable after a year
	  or so anyway (they'll have experience and referrees).

	- sighting personal ID. Trusted third parties don't work. My
	  experience with Thawte, them issuing a certificate to
	  conf.linux.org.au, without anyone actually seeing any 
	  identification documents has convinced even more so. A distributed
	  web of trust is far more robust and is well supported by Free
	  Software -- GPG and various interfaces to it.

	  Plus there are plenty of people in the community, especially
	  the Debian one, well versed in GPG use and abuse. Keysigning 
	  parties are like tupperware parties of decades gone by - except 
	  with different kinds of plastic.

Regards,
Anand

-- 
 `` We are shaped by our thoughts, we become what we think.
 When the mind is pure, joy follows like a shadow that never
 leaves. '' -- Buddha, The Dhammapada