This morning the Panic guys released Transmit 3.7 but rather than it being a major release, the release notes simply state:
- Fixed an encoding bug that could cause issues with both international filenames and WebDAV.
which seems a little anti-climactic, but so it goes. Still, a great app by a fine bunch of indie Mac developers.
Today I spent a good bit of the morning finishing up a few things on my first yearly review here at The Shop. One of the most disappointing things about the review was that I needed to create a development plan for the coming year. This makes sense then you're talking about a 27 year old kid - but by 47, if I'm not on the top of my game - and able to stay there, then I shouldn't be in this position. It's like asking Tiger Woods "Yo, Tiger - what are your plans for a better game next year?"
I'm not saying I'm the Tiger Woods of coding, but I also don't need to have a development plan because people are put-off by my intensity, and some are scared to approach me. While I don't see this as a wonderful asset, I have to say that the review pointed out that my skills, speed, quality of work, and communication are the best around. "Raising the bar..." and "Setting the standard..." were quotes from the review.
So yeah, I can be focused and intense, and this can put off some folks. I don't mean it to happen, but then again, it's not like it's a mystery, either. Anyway... I needed to finish up some stuff with the review. Tomorrow it'll be over, and we'll see what happens as a consequence of this with the bonus. That will be very enlightening.
Once I was done with that, I had to mess around with this web app I inherited, and once again, found how horribly built this code is. The original author decided that comments were unnecessary at the time, and subsequently was not able to help me with anything I needed to figure out. He's a nice guy... I like him, personally... but if I were his manager, I'd tell him that if he did this again, I'd have him on app support for six months. It's a disgrace.
So I had to struggle through this bad codebase trying to figure out what needed to be done, and while it was only about 3 hrs of work, it seemed like an eternity. I want to do a good job, but it's tasks like this that make me want to scream.
When I got done with that, I was able to spend a little time fiddling with Filemaker Pro 10 on the application for a friend's office that needed a few changes. Interestingly, the changes were very minor - the format of a date field, the change of a report field, the calculation of the rows in a table... pretty simple stuff. It took me no more than an hour and all the things they've been worrying about for months were fixed.
It's amazing how unimportant customer service is to some folks. The original developer was getting $80/month for this app, and they wanted a few changes. He probably could have changed them in 10 mins, but chose not to. Things kept getting nasty, and finally, he told them it's over. That's a first - the developer firing the client. But that's when they came to me.
Too bad. He's out $80/month and they get what they wanted all along. His loss.
One of the tools that has made the last year of my professional coding very nice is the H2 Database Engine. It's 100% Java, it's fast, it's flat-file or in-memory, and it's just as rock-solid as PostgreSQL. I love both of these databases, and while I can use others, if I get the choice, I'll use one of these.
H2 is amazing as an in-memory database, but it does file-system database as well. It's got all the things you'd want in a database: multi-version concurrency control, triggers, foreign keys - the works. It's super simple to use as well. Could not be easier.
Love it when we see things improving in it.
Today I saw that Björn was able to release MacVim Snapshot 51 with a lot of additions - including a Core Text renderer that should be the fastest way to render text on the Mac in Leopard (and Snow Leopard). While I haven't seen a lot of lagging rendering, it's always nice to see the best you can see when you're hoping around and scrolling in an editor - like Vim.
Again, Vim is probably my favorite tool for working with text, and MacVim is an excellent port to the Mac. It's really something you have to try.
Well... now that I'm an employee in a 300+ person organization, it's Yearly Review time. I had avoided it for about 8 years as a consultant, but there's no avoiding it in large-ish organizations these days. It seems that everyone is interested in getting feedback from the employee on their year, giving them feedback on the reality of their year, and then having them provide feedback on others around them. It's a process that's pretty standard with a few local modifications in most places I've heard about.
My question is this: Is it really a useful exercise, or is it just something to fill HR folders? I'm not convinced it's a clear-cut answer.
Certainly, the biggest advantage of these types of reviews is to cull out the under-performers. In this litigious society, every CEO and HR manager is trying to avoid a civil suit on some wrongful termination. So rather than worry about those lawsuits, they have these reviews. These form the basis of a written, regular, feedback of the employee's lack of acceptable performance. Then, when they choose to fire the person, they have all the written records to point to so as to avoid the lawsuit.
But is that really necessary?
Certainly, it's one way of protecting yourself against lawsuits. But so would formal letters of poor performance. In fact, that's how they did it 30 yrs ago. You were told you're not "meeting your numbers" a couple of times by your boss. Then you got a letter, and then you got fired. It's pretty simple, and it's pretty direct. There's a cause, and an effect. There's no reason to wait 3, 6, or 9 months for the annual review process - you let them know when it's appropriate. Period.
Even in this day and age, I've been taken aside and told "Hey, lighten up" from time to time on a particularly tough day when I've been grumbling a bit. I can accept the reprimand, life is like that. If you're working your hardest, I for one, am not surprised that you ruffle a few feathers. So, in general, I'm a big fan of the "as needed, when needed" approach to both negative and positive feedback.
You can't train a dog not to poo in the house by yearly reviews. Address the issues when they occur with the proper level of documentation, if necessary.
But for those overly cautious types... those worried about civil suits... I can see some value in the ability to document people that should be fired. But the downsides of the review process to nearly everyone else is staggering.
Each review process has several categories. I've seen as few as 5 and as many as 20. They are all just varying degrees of job component classifications. In that, they are all the same. So let's look at the typical employee first. They are a middle-of-the-road employee. They get their work done acceptably, nothing amazing, but nothing bad. They blend into the woodwork, as it were, and do their jobs.
This person may think of themselves as a A-quality person in some regard. Maybe they are right, maybe not. But what's the real value of telling them that they are a C-quality person? Does it make them work harder? Not really. They are a grown-up. They have made their decisions. Maybe it's their home life, maybe it's the time they need to spend in their bowling league, but for whatever reason, they are giving the employer all they are going to give.
So who gains by telling this person they aren't as good as they'd like to think? No one. The reality is not going to motivate this person - in fact quite the opposite. This is going to demotivate this person. It might even make them feel a little hostile towards the manager or employer. Why? Because as much as this person may be kidding themselves, they wanted to believe that their efforts - and everyone makes efforts from time to time, is seen and appreciated. This "honest" feedback may be honest, but it's not going to help.
The alternative is true as well. A person that seems them as a C-quality person is not going to believe they are an A-quality person. They have a bad self-image, and no amount of management praise is going to change their self-assessment.
In either case, this isn't really helpful feedback. It may be necessary for a lot of reasons:
useful reasons, all - but not really helpful to the employee. Face it, all these can be accomplished with a talk from the manager or again the old style letter.
Let's look at the case of a person that is exceptional in nearly everything but one activity. Maybe they don't communicate well. Maybe they don't interact with customers in the best way. Maybe they don't understand the business as well as others. But in all other regards, this person is a star. What about them?
This person has decided that one part of their profile is just not as strong as the others, and like many human beings, chooses to do something they enjoy rather than work on something they do not. So they pour more energy into the things they are good at, and leave the things they are poor at to suffer. What does the review process do for them?
Maybe it is news to them, but most likely, they know exactly what the truth of their skill-set is. They don't need to be told. It's almost annoying that their faults are brought up again at this time, when it's clear that in every other regard they are such an incredible asset to the team. They aren't going to get cut. But every year they are going to be told how great they are except...
If it's not news to them, then they fall into the last category.
In this case, the person knows how they are seen, what their strengths and weaknesses are, and has come to terms with them. They aren't going to be changing because someone fills out a form. They have decided exactly how hard they are going to work. They are either going to surf the web - or not, but no amount of reviewing is going to change their mind. This is who they are. Period.
To many, this person is annoying. In the extreme. They seem not to care that they can be better - no matter how trivial the work may seem to be to improve. They play to their strengths, and accept their limitations. Because of this, most of the time, they are very successful. When they aren't, they don't loose sleep over it. It just is.
For this person, the process is a joke. It's someone else's idea of a way to cover their behinds in case of the civil suit for wrongful termination. In the end, this person drifts through the process, accurately representing themselves, and the opinions of others, which possibly interesting, are not sufficient to sway this person from their course in life. It is, after all, their life, and they have purposefully chosen this path.
So I look at all this and think in the end, it's a waste of time. Talk to, warn, and fire those that need firing. For the rest, tell them when they do something good, and talk to them when they do something you wish they hadn't.
Anyone that has kids knows this. If you want good kids, talk to them. Don't make it a yearly process. It just doesn't work that way.
This morning I noticed that Hulu Desktop 0.9.8.2 was out. It was only a week ago that 0.9.7.2 was released. This could be part of the build-up to the stated plans by Hulu to start charging for content in addition to the ad revenue. While I'm not against someone charging for something, it will make me delete the application as I have no interest in paying for content like Hulu when I can get all the content I want in iTunes for a price.
It's their choice, but I have a feeling this is the motivation for the quick update.
No release notes are available.
Today I've been doing a fair share of heads-down coding on the Google Visualization DataTable in Java - to get it up to the current JavaScript API published by Google. I started with this update a few days ago, but I stopped at the filtering and sorting. Today I dug into the filtering and allowed either a Java List<Map>, or the standard JSON array of maps. To make this a little easier, I made a parser to take the JSON string and convert it into the Java List<Map> and then have the JSON version of the method call the Java version. Pretty simple, but the converter was a little tricky, but not too bad.
I was pleased with the filtering. It worked wonderfully, and while it's probably not the highest performance implementation, it's pretty good, and certainly good enough for what I'm going to be using it for initially, which is just filling out the API spec. But if there becomes a problem later, I can imagine a few ways to make it nicer, I just didn't mess with them, and so performance tests to see which is better.
But the really nice thing was just that I was able to shut the floor out, listen to some decent music instead, and focus on coding. The one thing I really enjoy. That was the treat I gave myself today - the pleasure of creating. I really don't get to do it enough.
I was reading Daring Fireball this morning and the following was a posting that hit me the second I read it. It's an interview with Cormac McCarthy:
WSJ: How does that ticking clock affect your work? Does it make you want to write more shorter pieces, or to cap things with a large, all-encompassing work?
CM: I’m not interested in writing short stories. Anything that doesn’t take years of your life and drive you to suicide hardly seems worth doing.
I love the passion expressed in that one line. Create something so utterly fantastic that it demands of you years of life or nearly every shred of your sanity. If you're doing less, then you could be doing more. And if you can do more, then you should do more.
I can't find who said it first, but this Life is not practice for anything. Do, and be, and follow your passion. And if you have none, see if you can find someone to infuse you with some. You'll be happier for it.
Big news from the alternative browser space - Camino 2.0 has been released and is out there for all. It's not my top-flight browser choice, but you need to have a bunch of them to test all the pages you make, and this one is nicely Mac-like with the Mozilla engine behind it. It looks like it's got a new tab overview - like Safari's latest pages feature. Could be nice, sort-of like Expose for the browser. We'll see if it's really useful.
This morning I noticed that DrawIt 3.10.2 is out with a nice little update list in Sparkle:
- Slightly updated user interface in the filters stack
- You can now click an effect in the favorites to add it, instead of having to drag
- The visible and locked symbols are now displayed properly in the list
- Adds Noise and Random Noise filters
Had to upgrade, of course.