[thelist] show of hands...

Ben Henick persist1 at io.com
Thu May 9 21:59:01 CDT 2002


On Thu, 9 May 2002, John Corry wrote:

> > Of course, it's not for everybody.
>
> Thanks Ben, I really appreciated it...

Appreciated *what* - my confirmation of your experience, in terms that
lean over to the perspective of the guy sitting at the other side of the
table?

> I try to get off-site subcontracting work all the time, never goes
> anywhere. If you were looking for a programmer, why would you hire a
> graphic designer/information architect. Or vice-versa?

That's one way of looking at it.

Let me go ahead and share my own experience.  [Erika Meyer also posted to
the site a narrative of her own along the same lines, in October or
November of 2000 IIRC.]

Agency No. 1 makes me a candidate, starts trying to line up work.  Three
inquiries come through:  one for QA (really usability testing), the second
for production artist, the third for a variety of JavaScript.  Folks get
hired for the first two before I get a client interview, the third I get
an interview but bomb after answering a loaded (and possibly illegal, but
of course I couldn't prove it in court because my rep wasn't a witness)
question honestly.  [What I do when I'm not on your premises is none of
your g-----n business as long as I'm a genuine asset to the project,
thanks.]

Agency No. 2, where my rep (as I learned recently from other quarters) has
a tendency to push the candidate at the front of his queue into the job at
the front of his queue regardless of whether or not the two are a good
match, asks me to interview for a CF development perm with a dotcom.  No
matter that I've got that agency's best-of-market scores in markup
testing; no matter that I prove competence as a copywriter, IA, and
production artist; no matter that CF is mentioned nowhere on my resume; no
matter that at that time I've only taken one dynamic site to launch as a
design lead.  Again, I bomb the interview after being faultlessly honest.
"Yes, I can do the job (so I say, knowing that CF is markup- and
Windows-based) but I'll need 90 days before I can start quoting
time-to-delivery without cringing, and without running the risk of 60-hour
weeks."  Oops.

Agency No. 3 puts me on a production/architecture-revamp gig without
registering me; the gig is supposed to be 6-12 months, but only lasts two
weeks.  I am put into a hinterland, 50 yards from the rest of my team - I
had considered doing an illustration of a tiki and pinning it to my cube -
before I've spent two hours onsite, and it takes them until the fifth day
to get me a workstation and a login.  At that point I get the Access
printouts and discover that the new IA STILL reflects the internal
organization in preference to customer priorities, and make the mistake of
saying so out loud.

Agency No. 4 has only gotten me one interivew, which was for a Netscape
server programming contract gig.  I'm ballsy with the interivew and almost
get invited to take it, but it's decided that they'd rather hang in and
see if they can get someone with bonafide server-side JavaScript
experience.  Shortly afterward my mother winds up in the ICU with
life-threatening cerebral trauma, and nothing else matters for several
weeks.  (Eh, it would've been a four hour daily roundtrip, anyway.)  The
same agency invited me last summer to interview for a state govt. design
lead contract, but I had to turn it down because I had a job at the
time... and their main Web client right now (also a gov'l org) is so far
into the red, it's pathetic.  But when I talked with my rep, it was
interesting... we talked about the situation for a while, I pointed out
some alternative solutions to problems the project director had mentioned
and passed on to my rep.  But at the end, it was "this is straight
production - would you be okay with it?"  Of course... I say, "Jeez, I
just wanna have steady work for a while" as I think to myself, "Ramen
noodles get old fast."

When dealing with these situations, it's frustrating... because I know
that the end client can - especially in this economy - find somebody to do
exactly {x} better than I can, regardless of what {x} is (with the
possible exceptions of copywriting - for which I have no credentials, just
clippings - or site optimization, which doesn't matter to big clients
anyway).

Where I see myself in a team context is as a PM, design lead, or
pinch-hitter... but 'presence' is a problem for me (as I mentioned
earlier), and design lead gigs are more likely to go to degree-holders.
That leaves pinch-hitting, which never comes up because PM's don't
consider it a viable skill.

And who can blame them?  Somebody who can hold state well enough to "fill
in the gaps" is also likely to be someone who will question, privately or
publicly, silently or aloud, the project manager's authority every time
the PM makes a bad decision... despite the fact that the line folks on the
team likely are kept as disconnected from the politics that led to those
poor decisions in the first place.  Taken in combination with the fact
that the utility guy will produce at a lower volume than the first-team
specialist, it's just not worth it organizationally, even if it shores up
the team's ability to meet deadlines.

For me personally, much of the problem has been my directness.  In
retrospect, I can't avoid the lesson that during the dotcom era I
should've talked big just like all the other bozos... though the reason I
didn't is that I'd've had a hard time looking myself in the mirror down
the line.  I hate being sold twenty pounds of s--t in a ten pound bag, and
refuse to do it to anyone else.  Even though it seems to be the way of the
world these days (which is one more respect in which the recession hasn't
improved things much).

But I *will* make damned sure that the site goes live on schedule, barring
gross stupidity on the part of my managers.  Even if I give myself a heart
attack in the process.  (Loyalty, integrity, and the inability to break
promises with a clear conscience have this odd tendency to show up in the
same list of traits.)

My own experience suggests that if you're a generalist and hope to have
any success with third party agencies, you need to:

1.  Be honest with your rep about your skillset
2.  Find a rep/contact who will be honest with you about her needs, in
    return
3.  Choose one or two specialties (producer, production artist, DBA, PM,
    and Flash developer would be the high-percentage ones) and agree that
    the third party agency should concentrate on those sorts of gigs in
    your case
4.  Prefer contract work to permanent work
5.  Be quiet as a mouse on matters outside of your role, no matter how
    badly it hurts

> It seems like the only real job I can successfully do (or s--t, even
> qualify for) is 'web swiss army knife'. I can do a whole lot of tasks in
> web development, but inherent to that broad a skillset is the reality
> that I'm not an expert at anything. S'ok with me...and I've yet to work
> for client who actually needed an 'expert' (or would be willing to pay
> expert rates), but sometimes I get sad thinking about how half-ass I am
> at everything I do...lol, that's just a mindf--k though :)

I don't think you give yourself enough credit.  (Which you hint at.)

Much of this, as I mentioned in my previous message, has to do with
presence and negotiating skill.

As a generalist working alone, you:

- eliminate the daisy chain
- serve as the sole node of accountability
- can leverage your strengths against your weaknesses, to an extent that
  only a top-notch PM can match... because you know yourself better than
  any PM knows the individuals on their team
- save the bookkeepers a boatload of work
- simplify onsite politics, which tends to reduce budget requirements
- are better equipped as a matter of course to take initiative
- can postulate and satisfy business cases without laying on a layer of
  bulls--t

While difficult to measure, these benefits can and should be billed at a
premium.  *nudge nudge wink wink*

> One interesting thing that's' kind of relevant to this topic that I
> experience recently:
>
> Web teams need to be either really small or really large to be
> profitable. One generalist can make money. 10 (?) specialists can make
> money. A team of 2-3 generalists is useless, too much time is wasted
> communicating and keeping everyone up to speed. Dunno' that's how it
> worked for me, YMMV.

YMMV is right.

One generalist can make money, if they've got a good grasp on client
service, good negotiating skills, and an ironclad committment to meeting
overhead.  (My gut feeling is that Shirley's probably best qualified to
discuss those things, but lemur seems to have 'em nailed down quite
nicely himself.)

10 specialists can make money, but the project in which that many people
are necessary is rare IMO... teams that big exist to satisfy egos more
than project requirements, unless the project is so well-planned and on
such a short timeline that ten developers can somehow pull it off
without eventually tripping all over one another.

Find me a Web project built by an *expert* team that requires much more
than two man years to build, and your next task will be to introduce my
jaw to the floor... my opinion is that if it's taking any longer,
somebody's not doing their job well, or the client's refusing to make
decisions and stick with them, or the scope is caroming and careening.

{insert reference to "Mythical Man-Month" here}

That commentary springs from experience - a couple of years ago, I and one
other developer built a seven-feature, user-customizable T&M processing
system in six weeks (800 hours total; Chris stayed on two weeks longer
than I did) despite daily interference and scope changes by the client
(who knows how to manage engineers, but couldn't lead a software or Web
developer to the fsckin' bathroom IMO).  Thus, the thought that a team
larger than four or five is necessary for a project, is kind of credulous.
Though in our case, it helped that I'm an oldtimer, and Chris is just
simply brilliant.

...And if the dev environment hadn't been such a bleedin' circus (by the
time I left, I was getting skylined on a daily basis - I knew I wasn't
going to get paid for the scope changes, and had become a deeply unhappy
camper) the total labor probably would've been cut down by 15-25%.

(Note to self:  Figure out how to avoid being nothing more than a
lightning rod.  This martyr crap'll be the end of you.)

A team composed of a small number of generalists is EXTREMELY useful.  The
problem is that the billings will be front-loaded... the tradeoff is that
such a team will catch their mistakes more during planning and
development, and less during testing.  The same is more-or-less true in
the case of a single hired gun, though the likelihood that such a
generalist will need to do a lot of testing is lower than it would be with
a full team.

Of course, the preceding statement assumes that the generalists in
question have figured out how to work smoothly together, that they know
when to stop critiquing and move onto the next phase of the project.
Which is hardly easy.


--
Ben Henick
Web Author At-Large              Managing Editor
http://www.io.com/persist1/      http://www.digital-web.com/
persist1 at io.com                  bmh at digital-web.com
--
"Are you pondering what I'm pondering, Pinky?"
"I think so, Brain, but... (snort) no, no, it's too stupid."
"We will disguise ourselves as a cow."
"Oh!" (giggles) "That was it exactly!"




More information about the thelist mailing list