Richard Stallman represents everything that’s wrong with IT’s image in the larger business world. The things he says are often insightful and informative, but his legendary lack of anything resembling people skills makes me want to plug my ears and denounce him as a crazy old nerd–and I’m not the only one.
As a result, he does more harm than good. Need an example? How about his condemnation of cloud computing, which he says will lead to “loss of control of data.” He has a perfectly valid point: as the linked article points out, users of Amazon’s cloud service had Wikileaks information deleted without warning or recourse because the provider decided it violated their terms of service. Believe me, the ownership of cloud-hosted content and the associated legal issues are going to be hot-button topics in the coming months and years, and businesses concerned with proprietary material are going to be watching and waiting with bated breath…to say nothing of journalists, musicians, artists and regular folks concerned with Google using their photos of Aunt Jane as advertising fodder.
But Stallman takes those legitimate concerns and wraps them in a layer of long-haired anti-social neo-hippie crazy that makes the rest of us shrug him off as just another loon. You want to take up a contrarian position simply because you don’t want to be associated with the guy.
I mean, he’s discouraging people from using the LOIC program against companies suppressing the Wikileaks info–but not because it’s illegal, immoral and does damage to the case of those supporting free speech online. No. He doesn’t like it because “it’s not open-source.”
Stallman just needs to stop.
I was just reading this article about the Airbus A380 that had an engine blow up (they landed safely), and this quote from the interview caught my eye:
“…the fuel dumping system had failed and we were about 50 tonnes over our maximum landing weight. In the Airbus and the A380 we don’t carry performance and landing charts, we have a performance application. Putting in the ten items affecting landing performance on the initial pass, the computation failed. It gave a message saying it was unable to calculate that many failures.”
I’m a little humbled by that statement; in my life as a web developer, I think the general approach to error correction has been “if that many things go wrong at once, reboot the server and start polishing your resume.” This is a great illustration of the need to properly consider failure modes when designing software–and to be aware of the consequences! Kind of hard to reboot an airplane…
This brief blog entry and comic illustrates the battle between the job seeker and the job filler: the signal-to-noise ratio is perhaps more lopsided in the job search market than anything other than a web forum filled with angsty teens. Networking is by no means a guarantee, but it gets your a lot closer to your target.
Business is people!
I’ve just finished looking through hundreds of channels, our DVR, and an on-demand video service, and I’m right back to the laptop. I can’t help but wonder if “57 channels and nothing on” is actually caused by too much choice rather than not enough? Custom content for all is just sophisticated market segmentation, but what if all that division ends up removing the interplay of elements that actually ends up making a better product? Consider soup: maybe you really, really like egg noodles, and somebody else really likes chicken, so you offer two separate menu items: noodles, and chicken. But maybe it turns out that neither is as good alone as they are together.
I think the industry has gotten good at determining individual elements that people enjoy: we know that some people like stuff blowing up, so we’ve got Destroyed In Seconds, and we know that some people like celebrities, so we’ve got Keeping Up with the Kardashians. But I think that as market segmentation gets more advanced, we’re going to be moving beyond individual tastes and going back to finding a perfect blend, a recipe that pulls in a broader demographic through the combination of individual preferences than by just creating many different shows. What if one single Cosby Show actually brings in more eyeballs than a dozen “niche” programs combined? Can you figure out how to achieve synergy by combining disparate market segments, by adding some explosions to your soap opera, or some romance to your sports?
Can you create something everybody wants to watch?
Who knows? Maybe sitcoms and variety shows will be the next big thing, all over again.
I’m semi-officially hitching my wagon to the Salesforce.com star, so I’ve decided to certify my in-the-trenches experience by taking the Certified Administrator exam on Wednesday. If you read about the value of certifications, you’ll come to one conclusion: the consensus is inconclusive (given, that first link is from 2004, and I would guess that certifications go up and down in value depending on job supply and demand). As big a proponent of education as I am, I agree that you’ve got to put classroom lessons to use; I’ve seen plenty of people with certificates out the wazoo who couldn’t tell a CD-ROM drive from a cupholder. But does that mean certifications are utterly worthless?
One point I think these articles aren’t focusing on enough is that a certification may not be a value-add feature in a candidate; given two otherwise equal candidates, you’re not going to hire both of them but pay the one with the certificate a bonus. Rather, a cert is a baseline requirement; it’s a barrier to entry, meaning that the candidate without the cert doesn’t even get to the next stage of the hiring process.
Think about it from the perspective of the gatekeeper: if you’re an internal recruiter or hiring manager at a large firm looking at 500 candidates, a certification helps you quickly winnow down the pile. Is it right and/or fair? Maybe not, but it may be reality. After all, while IT wonks and cube dwellers may be right in the belief that a certification doesn’t guarantee expertise, it does tend to require the certificate holder to have at least been exposed to the material. That can eliminate wasted cycles in the initial screening process, which makes the recruiter’s job that much easier, and reinforces the practice.
Bottom line? A cert can’t hurt–but it may not do anything more than open the door.
I think Google’s timing for Chrome OS (which is pretty much the talk of the Webs this week) is spot-on. Google is promoting the cloud from the early adopter/startup side of operations. In the meantime, folks like Salesforce.com are handling the big enterprises who are five years away from being able to contemplate dropping Office and Outlook for Google Apps, but who need to be (or want to be perceived as being) cutting edge enough to get onboard with this whole “cloud” thing the kids are talking about.
I’m pretty confident that the cloud is Web 3.0; I don’t think I referred to it quite as such when I talked about the iPad back in February, but the sentiment was there. This is just one more step towards that future.
The Brickskellar was probably the first bar I ever visited in DC. I discovered it when I was still expanding my beer horizons, and you can’t get much more variety than a place with over a thousand different options. Unfortunately, it looks like it’s closing down.
While I’m sad to see it go, keeping a bar/restaurant running for over 50 years is an impressive feat (although I was surprised to learn that the failure rate for restaurants isn’t nearly as high as the “90%” often held as conventional wisdom). And, I’m sure the owners are happy to finally be able to execute their exit strategy–if they are exiting, given that they’ve got another bar in a much more high-traffic Chinatown location.
The fire alarm in our building was going off when I woke up this morning. It’s not hooked up to the local emergency services system; you would still have to actually call 911 to get a fire truck to show up and investigate and/or fight an actual fire. That turns out to be a good thing, because all four times it’s gone off since we’ve lived here, it’s been a false alarm. We didn’t understand how this informal system worked until a former neighbor showed us how to hit the reset button–after conducting a tour of the premises for any obvious signs of a blaze, of course.
On the one hand, you might say that it’s a bad idea to allow just anybody in the building to determine that there’s no cause for panic. On the other hand, maybe it makes sense to admit that we’re doing exactly what the “expert” might do when he arrives: looking for obvious signs of trouble, and canceling the alarm. And if there really were a serious problem, somebody with inside knowledge–say, a person who lived inside the condo that was actually on fire–would raise the alarm a second time. I wonder if the same “system” might work in a business context. If just anybody could shout down an initial panic over a project in trouble or a customer threatening to walk, would that be a good use of common sense, or do you really need expert feedback?
I think that powers of ten serve as a great metric for dividing up companies. First, you’ve got a company of 1–a sole proprietorship, which can be anything from a hugely successful public speaker to a guy driving a cab to a student who decides that being a “consultant” is better than being unemployed. Then you’ve got 10, which is the first “real” company…you have responsibilities, as a CEO, to make sure that your people are taken care of, because at least 5 or 6 of them are depending on you for benefits and salary. 100 is a level that few startups even reach, where you’re bigger than small, dealing with issues of scalability and liability and HR, starting to get noticed by the big boys, and having to really, really worry about whether or not they’re going to just squash you out of existence. At this stage, either you stay a niche player forever, or you figure out how to get really, really big, really fast, before you get eaten. There are big decisions to be made around 100.
At 1000, you’re a firm, not a company. You’re thinking about putting your logo on the building you lease. You either own a space entirely, or you’re a big player in an even bigger space. At 10,000, your hiring and firing patterns can determine the life and death of small towns. You are a juggernaut. If you’re bigger than that, you’re almost too big to fail. You sell sugar water internationally, or determine the media habits of continents. You are a corporation-state.
Apparently, the best time to publish new blog posts is on Friday and Saturday mornings. However, there is about a four-fold increase in unique views if you publish more than once a day. I have also heard that two paragraphs is a good standard length for a blog post.
So, longer entries would probably do better on a Friday or Saturday morning, and frequently-published short bites will fill in the rest of the week. I’ll stick to that promise for the month of December; I realize that most of my writing on here is WALL OF TEXT, and it’s probably off-putting to some. What do you think? Short and sweet, or long and detailed? How often is too often? And is there a difference in best practices for business blogging versus personal blogging?