Tuesday, March 3, 2009
Where to go now
I usually don’t let my political reading bleed over into my technical writing, but this is something that I really think deserves to be shared with my fellow tech workers. Much wisdom follows. Read on.
In response to a diary by a recently laid off tech guy:
Where to go to now!
Here’s what I do, I go after small businesses with 5-10 employees and set up a monthly retainer for desktop support, system integration with internal and public facing website, design, building, management and infrastructure. I’m available most hours of most days. I try to make sure the revenue comes from as many sources as possible while still not sacrificing quality.
I would set up a website in your local area, get some cars printed at Postcardpress.com - 4 color - for $60.00 and start cold calling. Forget the traditional job resume. Unless it’s sensitive job, people will be more concerned with availability, reaction time, competency, likability and affordability.
I try to make it difficult for a small business to justify going off shore or using a company that has abused H1B employees by packaging as much in a retainer as I can, like website and mail, IT services, website building, troubleshooting. I’m able to spend a lot of time at home as the first thing I do is set-up a WAN between their office and their homes. They love that. Then I’m able to get in and do 90% of support by remote.
I learn all of the staff’s name usually by naming their computers and I treat each one like they sign my check. I follow-up and I also use my business experience to make suggestions on how to increase revenues and cut costs. That way they see that monthly as a investment that could yield big returns in productivity, revenues and taking the pain away with trying to figure where to buy which technology and why they need it.
That means getting the manual on their phone systems and their massive scanner copiers while discussing compliance with document and email archiving. These are things most IT people who have had a strict job description can do. But it’s hard work. Getting the clients and keeping them. Some you have to let go.
They generally hate to buy anything tech related by usually when they have a religious experience like a lost customer due to poor service, employee theft of data or all day on websites ( ha) or a massive systems failure and no back-up, suddenly they pick up your card and call. It usually starts off on a project basis. Show how useful you are and suggest a retainer as they call more frequently.
I usually find customers working on computers years old when they were fairly expensive, reluctant to trade up because they assume it would be too costly. Employees watch progress bars, spinning clocks and hourglasses. Everyone is frustrated. You solve that and they love you.
You’ll be surprised at how many small businesspeople need us but didn’t know they could afford us. They have stumbled along blindly for years. You’ll see some things that will make you shudder. They also don’t like to make calls for help to many places and hear the finger pointing especially if it’s offshore.
It’s a market that has been a bear to break for large IT businesses and a field of dreams that never can seem to be build for sellers of IT equipment. It’s there. Go after that and don’t look back.
You can easily build a hard barrier to entry if you like the client well enough. Also be open to bartering too for Doctor’s offices etc. They all need you. They just don’t know you are there. Most small contractor companies don’t like anything under 10 desktops.
That’s where I live. It’s painful some days and some days it’s a breeze. Start now and in a year you’ll be fine economically and in good spirits. If you lose a client you only lose a percentage of your income. If you lose a full time job, you lose everything.
There are arguments against what I do, but they aren’t as applicable now as the country struggles. A recession is the big excuse to get rid of the most expensive employees for short sighted firms. I saw this in the Bush recession of the early 90s. They end up hiring the people they laid off as consultants at three times the rates. This time it’s different. The economy won’t come back for a long time. Chances are you will be profitable and around longer than your old company.
Also offer your services on a hourly rate to your employer. Hand them a rate card and a business card like any pro would. Ask for referrals. You may be surprised.
You can also offer your services to SOHOs too. Plenty of people start at home businesses and they need you. I’ve had small clients call the local “geek squad” but most of these companies don’t offer up people that understand business. Just gadgets. Neither do the offshore UT companies.
Forget the arguments over immigrants and H1Bs. You need work. When the income comes back you can argue the fine points. These political arguments and accusations don’t put food on the table.
By the way, I think I’ll make it through this year with two fill ups for car. Can’t beat that. :-(0)
Extended comment from someone with the handle Dburn.
Tuesday, February 3, 2009
This is why we have teams
Just had another reminder of why we have, or at least used to have, teams to do various things in Information Technology. A good friend and colleague was stumped by a problem with their code and asked me to take a look. Of course his code is way more sophisticated than my pathetic attempts at programming, so I had a hard time following it (he’s one of those guys who can do in a single line what it takes me a block of 5 to accomplish).
It was your basic add uniquemembers to an LDAP group kind of problem. With a twist. To keep his app within licensing limits my co-worker had come up with this clever audit routine that checked on actual usage patterns so that he could remove those who didn’t need access to a particular thing any more. On the LDAP side the program would do an unqualified LDAP delete on the uniquemember attribute in each access group and then repopulate with an LDAP add that had all the members who the audit routine determined still needed the access provided by a group.
The problem was he kept getting an LDAP error 20, “Attribute or Value Already Exists”.
After checking the obvious (LDAP access controls, app user dn and password) with LDAP browser, we started talking. Looking over the code I noticed that he had some “static” group members hardcoded into the script: guys who would always be included in a group even if they didn’t make use of the privileges it conferred often enough to avoid being purged.
“Let me check to see if there are any duplicates in that list,” I said.
Lightbulb over the head moment.
Of course. If this static list were added to whatever was discovered by the audit routine you could easily wind up with a duplicate. If that happened the LDAP server would fail the add operation with an Error 20, because at least one of the dns being submitted for uniquemember was already in a slot before it on the list (uniquemember is a multivalued attribute, its values stored as an array, or unordered list). Once we figured this out it was easy to fix by making sure the list submitted contained only unique values.
Moral of the story? Sometimes it takes a team to get things done. A lone wolf programmer could easily spend hours, days, on a problem that a relatively quick discussion with a colleague could resolve. Two (or more) heads are always better than one. Although this may cause some consternation for the bean counters in corporate finance, in today’s incredibly complex world of information technology no man, or programmer, is an island. We simply can’t do it alone.
This is why we have teams.
Wednesday, October 22, 2008
Taking a step back: The Sad Story of My VMware Downgrade
A short time ago I wrote about upgrading my personal systems at home and work to VMware Server 2.0, and gave what I thought were some compelling reasons why it was A Good Thing ™.
Unfortunately, the one issue I haven’t been able to work around has been my inability to get any guest, whether upgraded or freshly created, to access a USB device. This began as one more frustrating problem among many, but after a week of trial and error has finally become the show-stopper that forced me to downgrade to version 1.0.7, which thankfully is still available in the product archives.
Downgrading turned out to be very easy. All I did was remove the 2.0 package with a rpm -e VMware-server and then reinstalled the old 1.0.7 package with a rpm -ivh VMware-server-1.0.7-108231.i386.rpm. Finally, I ran vmware-config.pl and accepted all the defaults — except for the location of my virtual machines — and I was good to go. A quick test with an old Windows 2000 guest showed USB connectivity was working again.
It was really a disappointment to have to do this downgrade, especially because all of the good things I said about the new version (like the web interface, for example), still hold true.
Maybe version 2.0.1 will be better. Hey, a guy can hope?
POSTCRIPT: After flawless reinstalls on my work laptop and home desktop, I experienced serious grief on my work desktop in trying to uninstall 2.0 cleanly. The main problem turned out to be that some processes, including one guest, refused to shut down. A kill -9 finally dispatched that demon. The next problem was that the earlier version installer refused to replace the already existing later version kernel module with its own. I got around that by installing the latest centosplus kernel (but could just as easily picked any other kernel at random and then uninstalled and reinstalled the one I really wanted). At that point the installer had a fresh modules directory to work with and didn’t complain.
Tuesday, April 8, 2008
Breaking News: The (Outsourcing) Morons Get It!
Well, nearly.
The New Economics of Outsourcing
So now Infosys, Tata and others are looking to meet U.S. companies demand for “nearsourcing”, due to falling savings from getting services out of the increasingly higher-priced labor in India and elsewhere. So, of course, they’re looking in places like Brazil, whose economy is red hot and getting hotter and so should maintain its low wages for another … 5 or 10 minutes, at least.
Given the weakening dollar and increasing unemployment (not to mention what we’ve always been told were the reasons why India made more sense than anyplace else for outsourcing — language and education) right here in CONUS (Continental United States, and old military term from the 1970’s when the DOD still thought its mission was to actually defend this country, rather than “our interests overseas”), you’d think they would take a look at some of the more impoverished areas right here — like the rural South and inner city Northeast, for example.
Oh, wait. It seems like the folks in India, who are obviously better at business than their U.S. customers, have actually been thinking along those lines:
And they’re going to make a fortune. As will U.S. firms that begin to finally clue in to the great potential for this market — like the Triangle’s own Rural Sourcing of Durham, NC (not that their website got any play at all in either of the above articles — although there’s a nice piece here that appeared on NPR back in 2005).
Sunday, November 18, 2007
Good reason to have onsite admins!
There’s an article up on Slash , Sun to Create Underground Japanese Datacenter. From the article:
Kurtz’sKompund writes with word of a Sun project in Japan, one that’s taking a somewhat non-standard approach to data center construction. To save on power, heating, and water costs, the consortium is going to be building their center in an abandoned coal mine. The outpost will be created by lowering Blackbox systems into the ground; estimates on savings run to $9 million annually in electricity alone.
The obvious (well, at least to the moderately intelligent people who tend to visit Slashdot) criticism of this “approach” showed up in the comments very quickly. This thread pretty much said it all:
Are they crush proof? (Score:5, Insightful)
by rueger (210566) on Saturday November 17, @06:55PM (#21393535)
(http://www.threesquirrels.com/)
The Blackbox containers are robust enough to withstand earthquakes, being capable of withstanding a quake of magnitude 6.7 on the Richter scale.I don’t know, but placing servers 100m underground in a place that routinely is hit by large earthquakes seems a dubious idea. The containers themselves may survive a quake, but what happens when the disused coal mine collapses onto and around them? Even if the containers and servers survive, will the power and data cables? If the tunnels collapse how will you get to and from the servers for maintenance?
**
Re:Are they crush proof? (Score:5, Funny)
by darthflo (1095225) on Saturday November 17, @07:20PM (#21393675)
Two possible outcomes:
1: Mine collapses, buries everything under millions of tons of rocks and stuff, Blackboxes and cabling survives, Sun market’s “the world’s most secure datacenter”.
2: Mine collapses, buries everything under millions of tons of rocks and stuff, Blackboxes and/or cabling gets scratched and/or really damaged, Sun hires Godzilla (this is Japan, where Godzilla’s big in, remember?) to smash away them rocks and free the mine once again.
*
Re: (Score:2)
by Dunbal (464142)
Not to mention the fact that coal dust is extremely explosive. I wouldn’t like to see a few sparks in there after a major quake. But then again I guess they know what they are doing.
*
Re:Are they crush proof? (Score:5, Funny)
by couchslug (175151) on Saturday November 17, @08:01PM (#21393945)
“If the tunnels collapse how will you get to and from the servers for maintenance?”Good reason to have onsite admins!
So go ahead. Outsource your data center to Sun!
Of course there are more than a few data centers for hire sited in places like Orlando and Houston, which have had a couple of close brushes with natural disaster themselves the last few years. It’s a difficult trade off, being close to the beach you love, or the oil billionaires you need to suck up to. Guess the cost of real estate in places like Montana and North Dakota is a little too steep.
For another really stupid idea, see this old post from Ranum.
The only thing that dampens the humor of it is that since 2004 (when it was first published) an awful lot of companies have (possibly irreversibly) piled most of their technology eggs in the India offshore basket (decimating the ranks of U.S. tech workers in the process).
Remember the riots last Spring in Bangalore? Here’s a headline didn’t make it into the mainstream media:
Calif. IT honchos were marooned in their hotel rooms’ during riots
which is a verbatim a reprint of a story in The Bangalore Blog entitled Silicon Volcano. here’s a snippet:
Rarely do seven days pass, without a head honcho of an international technology company or two, passing through Bangalore — and the week gone by, was no exception. A top executive of one of the world’s biggest players in the computer storage arena, as well as a key name in networking hardware, were more or less marooned in their hotel rooms for two days. Most of their engagements were cancelled, as the frenzy of violence and arson that followed the death of iconic cinema personality Raj Kumar, engulfed the city.
Prompting me to ask the same question Ranum does,
“What kind of morons do they turn out in business schools, anyway?”