Re: Admire the rentacoder.com
From: infobahn (infobahn_at_btinternet.com)
Date: 01/23/05
- Next message: Sergey I.Grachyov: "Re: Admire the rentacoder.com"
- Previous message: Matt Gregory: "Re: Usability/UI design resources?"
- In reply to: Sergey I.Grachyov: "Re: Admire the rentacoder.com"
- Next in thread: CBFalconer: "Re: Admire the rentacoder.com"
- Reply: CBFalconer: "Re: Admire the rentacoder.com"
- Reply: Sergey I.Grachyov: "Re: Admire the rentacoder.com"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sun, 23 Jan 2005 08:34:49 +0000 (UTC)
Sergey I.Grachyov wrote:
>>>>>Buyers, are you here?
>>
>>>>No, this group is primarily subscribed to by developers. This isn't a
>>>>for sale group, or a spam group, so stop doing it.
>>
>>>Are you moderator ?
>>
>>No, but usenet etiquette going back decades says you should abide by
>
>
> Just "No" is enough.
No, it isn't. This isn't a courtroom. Mr Howard is perfectly
within his rights to add information to help you to understand
why the lack of a moderator does not mean "free for all".
>>>Do you have any practise with discussed subject or other open auctions?
>>
>>I have been working on software and hardware development projects for the
>>last 25 years. During that time I have also hired many full-time and
>>contract programmers, with widely varied results.
>
>
> There is nothing about discussed subject or other open auctions.
Such subjects are not topical here on comp.programming, so
Mr Howard is right to avoid discussing them. Nevertheless,
having read his articles for a number of years, I have a
huge respect for his knowledge and ability in computer
programming. I have considerably less information about
your programming ability than I do his. It is, however,
almost certainly true that he has never bothered to
compete in the auction market you describe, for the simple
reason that he'd want to charge a realistic rate for writing
a high quality product. Consequently, he simply would not
survive in the low-grade world of auctions. (Well, not
without compromising his principles. I have no information
on whether he'd be prepared to do that, but I suspect he
would not.)
> You have hired people, who don't care about public rating.
> If your employee failed - just no needs to update resume.
Those of us with a conscience feel obliged to record our
failures on our CV, as well as our successes.
>
> You say "widely varied results", but there is no public info such as:
> March 2004. Randy Howard hire Mr.ABC to accomplish ...<project's description
> skipped>...
> Mr.ABC is greate coder.
> Will hire Mr.ABC again
If Mr Howard did indeed recommend Mr ABC, I would treat that
recommendation very seriously. This is because Mr Howard has
established a reputation for competence in this newsgroup.
If J Random Customer recommended Mr ABC, however, I would
find it far less compelling.
>>Hiring the lowest bidder on projects has never been satisfactory
>>during that time period. Rather the opposite.
>
>
> "lowest bidder" - that _you_ say.
Yes, he is saying that:
IT'S A BAD IDEA TO HIRE THE LOWEST BIDDER.
In fact, it's just a waste of money. The lowest bidder is likely
to have misunderstood the problem and the constraints. Furthermore,
he is unlikely to have a solid grasp of his language of choice.
If these things were not true, he'd bid higher. Someone who knows
his tools and his job is worth more than someone who does not.
> I have good working history, so I can bid above average. Yes "average" isn't
> too big.
Your judgement of the quality of your working history appears
to be based on the opinions of people who don't actually know
how to write programs. If they did, they wouldn't have hired
you for such small projects, but would instead have written
those programs themselves.
Since these people can't write programs themselves, of what
value is their assessment of your programming ability?
>
> This is always up to buyer: how to choose right coder.
> What factors buyer should be taken into consideration?
> Just price? Or, may be, rating and responses from other buyers more
> important?
Only if those other buyers know what they're talking about.
And they don't.
>
> "lowest bidding" is method to create working history at first stage.
With the rather wonderful exception of open source (GNU, etc), you get
what you pay for.
>>>When I place my bid -
>>> I foresee how much time I need to accomplish my work.
>>
>>Magic. Most people can't come close to doing that without decades of
>>experience.
>
>
> Most people doesn't have experience with open auctions.
> They are train other skills.
Like programming, for example. We have not yet seen any evidence
that those who provide programming services via auction are
skilled programmers.
> For instance - let's search for "resume lie" via google.
Well, we can if you like. I got 758,000 hits. So what?
The fact that other people lie on their CVs doesn't mean
that I have to lie on mine, does it?
>>Of course, if each project is tiny, perhaps that isn't
>>as much of a problem. I define anything that a single developer can
>>complete in a short period of time (a few weeks or less) as "tiny".
>
>
> I am not depressed by "tiny".
> This is good experience, good money (in total), good contacts.
I won't argue about the money, since "good money" is a very
relative term. But the experience of spending your time working
on tiny projects isn't necessarily good, in that it isn't
helping you to make a success of a career in an organisation
working on very large systems. The sorts of hacks you can get
away with in your tiny projects simply don't scale well. As
for your contacts, we've already established that these are
people who don't know anything about programming, so I would
question their value.
> Even for "tiny" project you can spend a few week or more
> to find right person.
> But you can form your projects as request for bids,
> publish it and receive many propositions same day.
Why not just hire an expert for decent money? It's likely
to prove cheaper in the long run.
>
> Ah, I forget - you will never do this.
> Because you can't check: Do you obtain software legally or not.
IPR legality is indeed a good reason to avoid auction-based
programming services, but it's not the only reason.
> May be I should establish relations with your competitors,
> who know how to perform this check.
Oh, it's easy enough to find out whether the lowest bidder
violated IPR rules. Just hire him, pay him, and then wait
for someone to sue you. Is that what you had in mind? Some
people might call that a high-risk strategy, compared to
spending a couple of days hacking out a solution yourself
instead of hiring someone.
>>I could create 433 hello world projects within the hour.
>
> Good luck to find some buyers for your creative works.
Sadly, he would have few problems.
>>Numbers like that don't convey anything.
>
> That number will illustrate: system works fine.
By the same token, it is a well-known fact that horse manure
is good for your digestion. Trillions of flies can't possibly
be wrong.
Numbers on their own don't illustrate /anything/ properly.
You need semantics, too. Let's see what that 433 tells me.
We'll assume (in the absence of information to the contrary)
that you're halfway through your threescore years and ten.
We'll further assume that you've been writing these programs
since you were 22 (to allow for a university education).
So you've had 13 years. But everyone needs a holiday, so let's
assume 4 weeks a year, or 52 weeks all told. That knocks
it down to 12 years. We'll further assume that you spend
4 weeks a year trying to keep up to date in your field.
So we're at around 11 years now. We'll assume you work
for 8 hours a day, 5 days a week, so your total available
time is 11 * 52 * 5 * 8 hours, which is 22880 hours.
Dividing that by 433, we get a little over 50 hours per
project.
So /if/ you're really that old, and if you've really
spent that long, then your average project time is
just over a week. If you're actually much younger,
then clearly it's much lower. These statistics suggest
that few if any of the 433 programs were large ones.
>>It's just a statistic, and you
>>know what they say about those. What is useful to know is the
>>content, how similar they were, how different they were, and what
>>concepts and methods were employed. That can't be conveyed in 4
>>words. It would take rather a long time to cover that area if those
>>433 projects represented a large amount of unique development work.
>
>
> "areas" of other coders available too.
> Someone who will employ - can spend time to "cover that area".
I think you've misunderstood Mr Howard's point, which is that
if you really have written 433 different programs (as opposed
to, say, ten or twelve programs about forty times each with
only mild differences between them), then it would have taken
you much longer to do and would take much longer to describe.
>
> To save buyers time: If coder has been able to fulfill N projects -
> he can accomplish N+1 project as well. Please, see my "N" above.
You will understand the falsity of that statement when you can
find the flaw in the following proof that all integers have the
same value.
Statement A: "All integers in any group of n integers have the
same value." (The word 'group' here is used in the loose sense
of 'collection', 'bunch', basketful', not in its mathematical
sense. No, that's not the flaw.)
We seek to show that Statement A is true for all n. To do this,
we will use proof by induction. That is, we will prove that
A(1) is true, and then prove that, if A(n) is true, then A(n+1)
is true as well.
In any group containing only one integer, that integer must have
the same value as itself, so we have proved that A(1) is true.
Now we need only prove that, given A(n) is true, A(n + 1) is also
true. If so, then we can construct an induction ladder from 1 to
infinity.
If A(n) is true, then every integer in a group of n integers has
the same value. Let Z be an arbitrary group containing n + 1
integers. If we can show that /any/ two members of Z must have
the same value, then we're done.
Let Zi, Zj, Zk be arbitrarily-chosen members of Z.
Consider the subgroup Y, which comprises all members of Z except
Zi. Clearly, Y has n integers, so all members of Y have the
same value. Y contains Zj and Zk, so Zj = Zk.
Consider the subgroup X, which comprises all members of Z except
Zj. Clearly, X has n integers, so all members of X have the
same value. X contains Zi and Zk, so Zi = Zk.
Two things that are equal to the same thing must be equal to
each other. Since Zi = Zk and Zj = Zk, Zi = Zj, i.e. any
two arbitrarily chosen members of Z (with n + 1 members) are
equal.
We have shown that, if A(n) is true, then A(n + 1) is true.
We have shown that A(1) is true. Classic proof by induction.
All your number are belong to us - unless you are good
enough at mathematics to spot the problem. If so, you
should have no problem understanding why this...
> To save buyers time: If coder has been able to fulfill N projects -
> he can accomplish N+1 project as well. Please, see my "N" above.
...is not true.
>>>1. Coder has deadline. If coder can't accomplish project before
>
> deadline,
>
>>>buyer receive full refund. And coder gets low rating.
>>
>>Well, that may make the buyer feel better for about five minutes. But
>>ten minutes later he gets fired for missing the deadline for a project
>>he was responsible for delivering.
>
>
> Just 10 minutes for testing?
I hope you focus primarily on customers who speak your native
language, since there is less chance for misunderstanding.
I also hope your native language is not English. The alternative
is unthinkable.
> I take this list from Internet. Ideal projects for outsourcing:
I have no problem with outsourcing. I just don't think it's
a good idea to hire Jason from Leeds to do it, if doing so
would distract him from his homework.
> There is the buyer, who can send complain to RAC's.
> And coder will lose his rating.
That's no comfort to the buyer if he lost his job because
he missed a deadline.
> If you are ever act like RAC's coder - you should know:
> rating much more important when money.
And that rating is worthless unless it's provided by people
who know what to look for in a programmer. He may be surprised
to hear me say this[1], but I have the greatest respect for Mr
Howard, and I have good reason to believe that he is in a far
better position to assess a programmer's skills than J Random
Customer on an auction site.
[1] Randy: I *know* it wasn't that funny, okay? I was tired,
and you know what happens when tired people do Usenet...
- Next message: Sergey I.Grachyov: "Re: Admire the rentacoder.com"
- Previous message: Matt Gregory: "Re: Usability/UI design resources?"
- In reply to: Sergey I.Grachyov: "Re: Admire the rentacoder.com"
- Next in thread: CBFalconer: "Re: Admire the rentacoder.com"
- Reply: CBFalconer: "Re: Admire the rentacoder.com"
- Reply: Sergey I.Grachyov: "Re: Admire the rentacoder.com"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
