Is it time for a unicode update of std.uni?

H. S. Teoh hsteoh at quickfur.ath.cx
Tue May 5 14:36:06 UTC 2020


On Tue, May 05, 2020 at 11:33:31AM +0200, Robert M. Münch via Digitalmars-d wrote:
> On 2020-05-04 23:03:15 +0000, Mike Parker said:
[...]
> > BountySource works the same way
> 
> Which is not a good argument... IMO then there process is broken too.
> Just took a look at the site:
> 
> * many old bounties => system doesn't seem to work very good
> 
> * no information about AVG time from posting a bounting until done =>
> IMO a very critical KPI
> 
> * no information about AVG bounty size in relation to time until done
> => Again, cirtical KPI
> 
> >  -- you put the money upfront and wait for someone to do the job.
> 
> Which is playing lottery and not really a way to move things forward.

I don't disagree, but you also have to be realistic that this is a
volunteer-based open source project where, unless you're willing to
donate enough to hire somebody to work on something full-time, you can't
really tell people what to do. (Well, you can, but there's no guarantee
they'll listen to you. :-P)  This is both the blessing of an open source
project -- you can sometimes get talent that produces high-quality work
for free -- and also the curse: sometimes there's something that ought
to be done but nobody's interested to do it.

The reason things are structured this way is because it's not geared
towards an employer/employee (or employer/contractor) sort of scenario
where you offer X$, and somebody signs on to fulfill that work.  It's
more a volunteer/volunteer scenario, where the money you offer is more
like an incentive for somebody to step up to do the job for the reward
if they feel interested and the reward is commensurate with the effort
they think they'll need to invest (note: perception here is very
important, it may not correspond with the actual effort/cost). It's a
buyer's market, not a seller's market, unfortunately.

IMO if we really want to get things moving, esp. on important but
"technically boring" issues that nobody shows an interest to work on,
what we need is to actually raise the money to *hire* somebody to do the
work so that they are obligated under contract to do it.  As long as
we're dealing with volunteers, nothing can be guaranteed -- there may be
nobody interested to do the job, or people can lose interest midway, or
they just don't have the patience/persistence to push things through to
the finish.  The only way to get *guaranteed* work is to hire somebody
with an actual contract that obligates both parties.


[...]
> I even don't have a clue how many from the community would be willing
> to work on a contract/bounty/you-name-it. But if these people are not
> in the community where else should I search?

If it's really that important to you, maybe you should be posting an ad
in the job market.  The problem with volunteers is that they can choose
not to take the job, then there's really nothing that can be done about
it.  Hire someone with a legally-binding contract, then you can secure
some guarantees.

And I don't mean any of this in a dismissive way, it's just the reality
of the situation.  It's just the nature of volunteer-driven work: the
interesting things tend to get done, but the boring (though no less
important) work tends to fall by the wayside.  This is why most
open-source projects have great code but poor documentation: because
coding is more interesting and attracts more volunteers, but
documentation is boring and nobody wants to do it.  Unless you can
generate a high level of interest in something, chances are it will be
sitting there for a long time.  I don't see any other way around this
except to hire someone legally bound to fulfill their end of the
contract.  Once you have a legally-binding contract then you can
obligate them to do the work, even if it's completely boring and
uninteresting. Otherwise all bets are off.


T

-- 
They pretend to pay us, and we pretend to work. -- Russian saying


More information about the Digitalmars-d mailing list