- Feature Articles
- CodeSOD
- Error'd
-
Forums
-
Other Articles
- Random Article
- Alex's Soapbox
- Announcements
- Best of…
- Best of Email
- Best of the Sidebar
- Bring Your Own Code
- Coded Smorgasbord
- Mandatory Fun Day
- Off Topic
- Representative Line
- News Roundup
- Editor's Soapbox
- Software on the Rocks
- Souvenir Potpourri
- Sponsor Post
- Tales from the Interview
- The Daily WTF: Live
- Virtudyne
Admin
To elaborate: startups do not succeed because they magically manage to find a single person to perform the tasks of a whole IT team. They succeed because they manage to prioritise those tasks which must be done right, do without the others, and find the right person to carry out the project. But, of course, it requires an actual thinking effort from the founders.
Admin
I agree, it is a risk. Being happy in my job and older (40's), I wouldn't do it.
However, I did go for something eerily similar in my late 20's and it paid off well, in both income and future opportunities. A lot of great things do get accomplished by the young and naive while us old farts view it as crazy.
I wasn't there, so I don't know if there was more WTFery than the story indicates. This was perhaps a below average amount of WTF for most start ups.
It would be a better story if, in the end our would-be superhero was actually offered the job and refused after hearing about a ridiculous salary with the promise of stock options and bonuses.
Admin
I'm going to assume that "transfer the servers" and have "no outage or downtime" didn't mean "set up a second set of servers, get them up and running, the do a switchover".
Why waste money on a second set of servers when the first are working perfectly find and have already been paid for?
Picture the scene, one week later: IT guy barreling down the freeway, servers tied down to the back of a pickup-- tied into a pile of batteries, slowly draining...
Admin
"Postscript: Two years later, Martin released the first version of Hoopla to the rest of the world..."
Admin
Probably. I agree that having challenging needs isn't automatically a problem. As I indicated, it's all about the amount of support you're going to get from the powers-that-be. You can take on mountains if your boss allows for occasional "failure" (e.g. the IBM "Fired? I just spent $10 million educating you!" - the more challenging the problem, the more failure should be allowed). If your boss requires everything to come in on time and under budget, though, they better all be softballs.
Unfortunately, it's hard to gauge how supportive someone will be in the interview process. Just as you put on your best employee face for the interview, they put on their best boss face. More often than not, it's a gut check to see if things feel like it's going to work out - nebulous things can be as important as identifiable ones in figuring out support. (I came away with "bad vibes" from one place I applied, even though everything pointed to them being up-and-coming. I went elsewhere, although I couldn't identify any particular thing that was off. When they imploded and the specific horror stories emerged, only in retrospect I could identify the little things that pointed to the brewing problems.) I took the fact that he submitted it to TDWTF as evidence that his gut-check failed, despite the absence of any (post-editing) specific issues presented.
Admin
... and underpaid :)
Admin
Admin
keep dreaming
Admin
Sounds like you work at a company like that already judging from your comment.
Admin
This is the kind of job you run from. Asking one person to do the work of several is one thing... challenging and something a person might want to take on. However.. if the leadership thinks it's a good idea to forgo mature databases (even the open source ones) in favor of a homegrown solution as the backend to their core product... well there's no getting around that kind of mistake in judgement. The fact that the developer of said database wasn't around during the interview is also a red flag in my mind. Companies shouldn't be relying on superheros to operate. If the founder had written the core product that would almost be better.
Admin
Admin
They frequently do (work accidents).
But more likely is that the burger shop is chosen as a location for a rocket launch and then the return fire of the Israelis will get it.
All-in-all it would be better to open the halal burger shop out of range of most Hamas rockets, like Tel Aviv or Ramallah, rather than in Gaza or Sderoth.
Admin
Isn't going to happen. To explain this with brutal simplicity, nobody, not even Monty Widenius, has written an ACID complaint SQL database from scratch. So what kind of POS do you think it is? And how, given that it is going to have no facilities for replication, sharding and the like, are you going to do the migration without interruption of service.
The correct answer is "You need a team of at least 5 to do that. If you take me on I will recruit and manage the team for you, and I will get some really competent people to look at the database and advise on what to do about it."
They won't give you the job, but when it all goes pear shaped someone might just give you a call.
Admin
Shoveling cowshit in a barn on a 100 degree day in Wisconsin humidity.
Admin
Exactly. That's why it was stupid for Microsoft to create SQL Server when Oracle already existed. And it was stupid to create Oracle DB when DBase already existed. And it was stupid to create DBase when IMS already existed. And hey, what idiot decided we needed a "database", when flat files and VSAM already existed? We already had ways to store data on a hard drive! All this "database" stuff was just reinventing the wheel.
For that matter, why do we need hard drives? Just because punch cards weren't invented here, you figure you need to build your own storage device? What a waste of time!
If everybody would just use existing technologies instead of trying to improve them or make somethiing new, we would be SO much better off. We'd all be living in really nice, clean, shiny caves.
Admin
Huh, I join with those who say, "Sounds like a fun job to me." Setting up a network and migrating the servers are one-time tasks. With no indication of how much time they were willing to give to get these things done, it's impossible to say if the demands are reasonable or not. If they expected him to do this on his first day, yes, crazy. But nothing in the article said that. If he had five years to get this done, it would be a pretty laid-back job. I'm guessing the real expectation was somewhere in the middle.
Once everything is up and running, providing network support and admin support for the company's app doesn't seem impossible for a single person.
And suppose it didn't work out. So he finds another job. It's not like his money is on the line if he can't pull it off. Okay, if the company goes broke he might miss a paycheck or two, which hurts but is probably not a disaster. When he applies for his next job, I presume he will say that the company failed for reasons beyond his control -- no one's likely to blame the network guy for a company going bankrupt -- and he talks about his successes while glossing over his failures. Just like we all do on job interviews.
Admin
Probably not...a company that had any shot at all of becoming the last one standing in a situation like that would not be trying to fill half a dozen positions with one person as they went into production.
Admin
SQL Server was created when Oracle already existed. DBase was created when Oracle already existed (the software already existed even if Oracle company wasn't the owner).
Mixture of irony and confusion again detected.When hard drives were invented, the usage of punched cards for data processing was something that was indeed "invented here" by the same company. During World War II the company boasted of the way its punched card equipment helped governments use racist profiling to track down citizens, on more than one continent. Now, although the company no longer makes hard drives, their invention is used the same way.
Admin
MS didn't create SQL Server, they bought/licensed an existing product, then they made it a front-line product with a team supporting it. These jokers had some guy write a data storage engine as a sideline instead of just using postgres or something.
Admin
Well, of course my post was not intended to be a history of computer storage, I was just trying to make a point about the "re-inventing the wheel" slogan, so I admit, I didn't do any research but wrote off the top of my head.
So I just checked and you're right: Oracle DB was actually published in 1979 while dBase came out in 1980.
Who invented punch cards is a more complicated question. Punch cards were used at least as early as 1801 on the Jacquard loom. Herman Holleritch created the "modern" punch card for the 1860 census. I was not aware until today that Hollerith's company was eventually merged with several other companies to become IBM. IBM is generally credited with inventing the hard drive in the 1950s. So depending on definitions, you could say that the punch card and the hard drive were invented by the same company -- if you don't give Jacquard any credit and you count a company that was absorbed by another company as the same company.
Of course, all of this is irrelevant to my point: that it's often a good idea to "re-invent the wheel".
The analogy itself is a curious one. Companies can and do "re-invent the wheel", literally, all the time. Is there just one kind of wheel, used on every wheeled vehicle in the world, and produced by one company? No. A wheel that works well on a children's tricycle is not necessarily ideal for a locomotive. A good wheel for a Matchbox car is probably not a good wheel for a full-sized car. Indeed, even similar vehicles often have different wheels. My pick-up truck does not use the same wheels as my convertible, and I'm sure that other pick-up trucks and other convertibles use wheels different from either.
For this and many other products, companies regularly struggle with the "make or buy" decision. "Buy" is not always the right answer. I used to work for a company that made kitchen applicances. At one of our factories we made our own nuts and bolts. Why, when we could have bought nuts and bolts from numerous outside suppliers? Because the company carefully analyzed the subject and concluded that we could produce them more cheaply than we could buy them, and we would then not be hostage to someone else's supply problems, labor disputes, etc. At another factory we bought nuts and bolts, because the quantities we needed were smaller and it was not economical to produce ourselves. We considered having the factory that made its own bolts also produce bolts for the other factory, but shipping costs, time delays, etc made this impractical.
Manufacturers routinely conclude that, while there are products on the market that meet their general requirements, that they have specific needs that make producing their own a better choice. It may be cheaper to make your own than to buy. You may have higher quality standards than potential suppliers. You may not be satisfied with their customer service on a complex product. Your needs may be slightly different. Etc etc.
Without a lot more detail than was in this article, I can't say whether making their own DBMS was a smart idea. I'd tend to think probably not, but it's not obviously stupid. Bear in mind that if they're making their own DBMS to support their own product, they probably do not need all the features found in a commercial product. They only need to include the features that are used in the app that it is supposed to support.
Admin
Nice try Martin! How's that... um.. "Director" role working out?
Admin
Why are programmers criticized for reinventing the wheel when car manufacturers are praised for it?
Admin
In the real world the more likely scenario is something like this:
You get assigned more tasks than you can accomplish, expected to perform herculean feats and end up putting in 80 hour weeks to appease your boss, who cusses you out every time something doesn't go perfectly. The whole office treats you like a slave and you can't get anyone hired to help you because they want to keep IT "lean". When the time comes to ask for a raise, you're laughed out of the conference room. Eventually you are let go for "performance" reasons and forced to collect unemployment, going majorly into debt and maybe losing your car and home.
Admin
Well, you are the expert on "the real world". Just reading your comments in the roof-caving article is all the proof I need of that.
Admin
Since when do developers get founder stock for doing their job well?
I thought upper-management/venture-capitalists were way too greedy to allow such compensation to the those who make their wealth maintainable.
Admin
I had a somewhat similar position though more than one of us goons in the squad. I was a network/server admin. Servers were windows, linux, Solaris, and Mac. Had to manage tape backup of the SAN, email, webserver, a couple small HPC clusters, wireless and wired routers/switches, NIS, LDAP, and occasionally had someone from somewhere random in the world come in for desktop support. Particularly rememberable was one case of Japanese Win Vista trying to get me to setup wireless security and he didn't understand english and I couldn't figure out his keyboard/used windows > XP at the time.
In short companies often don't have budgets for experts in particular areas (for example we could have used a real storage administrator (we had about 300TB of spinning disk and 1PB of taped storage and daily sustained read/write to the SAN of about 2Gbps (and a lot of it was writes)), a security/authentication expert to handle the LDAP, NIS, firewall side of things) etc). What they got was me: a software developer/DBA. Nice :)
Admin
No no, if we're doing imaginary happy endings, how about he takes the job, but the next day wins the lottery and spends the rest of his life living a very fulfilled life with lots of coke n' hookers :)
Admin
a quote i read in a gaming magazine: "the biggest problem with the software industry is the hardware industry." newer computers need newer programs to keep up...