A brilliant management two-for-one

There are two excellent ideas in this posting on ZDnet, aside from the observation that you can scale Ruby on Rails if you avoid hitting the disk farm for static content:

1) Have a team devoted to rapid scalable prototyping.  (LinkedIn Light Engineering Development)

2) Use free apps for both proof of concept testing and marketing.

Read and enjoy.

Ruby on Rails: scaling to 1 billion page views per month by ZDNet’s Dennis Howlett — While a lot of attention has been focused on Twitter with questions about whether Ruby on Rails scales, LinkedIn has been quietly running a RoR application on Facebook that is beating down around 1 billion page view per month. Bumpersticker, a relatively trivial Facebook application that allows you to create a cartoon that you can [...]

Best,

Dak

Canada Revenue Agency extends e-filing deadline to 06 May 2008 due to website submission problems

That’s today’s news, and I ran into it myself. All of the Canada Revenue Agency sites are Sllllllooooooooooow today. Moving the filing deadline off is the Right Thing To Do from a tactical standpoint. However, you still have to pay any owing amount today. (PUBLIC SERVICE REMINDER: do that through your BANK, the government’s website is too slow to look this information up there!)

This problem does raise some interesting questions.

Did they do volume testing, optimization, or load balancing? (Do you?)

Are their pages set to gracefully degrade in the face of large loads (clearly not; for the most part, the actual HTML (XHTML, a tip of the hat for a nicely structured page) loads, but the linked images do NOT), and the pages are set not to cache, even for purely presentational, ‘nothing to see that has changed here’ pages. (Do yours?)

Moreover, when you actually try to submit the ‘netfile’ .tax file, the page doesn’t load. The announcement that the deadline has changed is on a separate news page, that’s behind a submit button, and it too takes forever to load.

Wow.

This is a solvable problem. Anyone care to weigh in on opinions as to how to solve it? I’ll do a follow-up summary and share some of my own thoughts later in the week.

Have you tested this sort of load with YOUR application? Does it matter?

(In my strong opinion: yes, it does matter, and you’d best be able to prove that your projected load is something you ARE handling, every day. You better also understand how your system will degrade under unexpectedly heavy load…like a million people clicking on the submit button multiple times because it didn’t respond fast enough

Best,
Dak

Getting cranky about “IT Policy,” and improving your password practices

I just read a helpful article about tuning organizational password policy but I’m afraid it rubbed me the wrong way.

What it says is helpful and mostly good practice, but it fails to address the problem from the perspective of the users, and does the usual “well, this will be a pain for the users, but it’s good policy, so we’re recommending it,” which is one of many reasons why people hate IT departments. (I say this as a seasoned IT professional, and I hate us, too.

IF you notice that YOUR passwords violate any of these rules, chances are that they are already broken. Change them now.

To all password users, everywhere: Make your passwords unguessable as best you can. If someone guesses your password, change it. Corollary 1: Since you can’t know if someone might have guessed your password, change it from time to time. If you feel that you have to make a list of passwords, make a list of reminders, not the actual passwords, and keep it safe (not where someone can look at it without you knowing about it). More detail below.

HOW:

Investing for retirement: results of a large simulation suggests conventional wisdom on asset allocation is wrong

The New York Times has published an article summarizing a very interesting study: based on market volatility, the notion of shifting from equities to bonds as we age may not be as useful as has been thought. Worth reading.

Best,
Dak

Free for a limited time: how to keep raccoons out of your green recycling bin

Toronto uses a ‘green bin’ system to recycle and compost organic waste. Toronto also has a highly active raccoon population, which quickly tips, opens, scatters, and eats the leftovers in the green bin.

There are many solutions to this problem. Without further ado, here is mine, and a bit of description of the design space.

The requirement: keep the raccoons out of the recycling.
The solution space: cheap, people-friendly, does not require locking the bin in an enclosure, since many people do not have a convenient place for the enclosure.

Typical solutions involve locking straps, multiple screws, and some sort of buckling mechanism.

Mine is one loop of Velcro One-Wrap, and a screw. Tools needed, a screwdriver, a sharp knife point, and a drill or small flat screwdriver bit suitable for making a pilot hole in the body of the green bin. (A Leatherman tool will frequently have all of these. I favor the ‘Juice‘ line for everyday use.)

1. Take the roll of One-Wrap. Hold the end of the Velcro in your hand. Wrap it around your hand until it completely overlaps on your palm.
2. With a sharp penknife or utility blade, cut a 5mm slit in the velcro so that you can pass a screw through it, just below the overlap region. You will use this hole to securing the loop to the body of the green bin with a screw, below.
3. Push the screw through the velcro, so that the head is inside the loop
4. Place the velcro loop on the bale of the green bin, and close it. Use the sharp end of the screw to mark the body of the green bin.
5. Using your drill or small screwdriver bit, make a pilot hole for the screw
6. Drive the screw into the body of the green bin

Notes on using this device:

• The velcro must be firmly and fully coupled on the overlapping part to keep the raccoons out
• When you take the bin out on the morning of recycling/garbage day, remember to undo the velcro
• In practice, the loop lasts for about a year

Feedback, requests for clarification, commentary, kudos and complaints are all welcome.
I’ve thought about illustrating this posting; please let me know if you think that would be valuable, or would like to do the art for it!

HOWEVER, I make no warranty or claim of suitability, as I’m NOT selling this to you. It does work for me. It won’t work for you. Seriously, you’ll lose a load or two of recycling to the raccoons because you’ll forget to close it properly, but it’s radically better than nothing, and very inexpensive.

As a helpful commenter notes, below, there’s also a good commercial solution (http://www.raccoonsolutions.com/).

Regards,
Dak

The Free Economy and Your Business

I recently had a conversation with Nathon Gunn (CEO of Bitcasters) and this article came up. Free! Why $0.00 is the Future of Business.

Why you should read it: Not everything is going to be free, and at least 80% of the readers of this blog haven’t thought of all the business models that Chris Anderson (Editor of Wired, author of The Long Tail) lists in this article…and frankly, it’s the best and fastest overview you are going to see on the topic. If you think that every free business model is the same as Google, Yahoo! or broadcast TV, you are missing something.

Definitely worth a read. Why you should listen to me: for you old-timers on the web, remember HoTMetaL PRO and HoTMetaL Free? That was SoftQuad, and I was there. And, frankly, your competitors ARE thinking about this, if you aren’t, and they’ll eat your free lunch.

Best,
Dak

Tieing it all together [tips, opinions, musings]

So, what, on earth, do all these postings have to do with each other?

Like you, I am a “person in the loop” system. I observe, tune myself to what’s going on, act, observe what happens, and repeat. What I’m sharing with you is what I’m currently observing, that’s “on the fringe” that is my “reading outside of my area.”

I personally view reading outside of my area as being critical to success. We have to be able to bring all kinds of ideas into play when we are looking for creating consulting and management solutions, and bring them up in a heartbeat.

The challenge is maintaining focus while doing this. One solution is setting a limited time budget, and a low energy time, when other, more critical things need to be done, so that it adds value rather than rationalizing distraction. It’s also a heck of a lot of fun, which is all the more reason to put it as a ‘time reward’ or ‘play period’ with limits.

Best,
Dak

High-speed digital camera

David Pogue just reviewed the new Exlim EX-F1.  This is another interesting data point for the ‘cheap and fast’ versus ‘high-end cutting edge.’

In my opinion, it’s both:  the cheapest digital high-speed camera, but high-end for a ‘digital camera’.

But what’s really interesting about this camera is that it changes what’s possible for analysis (product development, amateur and professional athletics, fault diagnosis).

Have a look, and weigh in!

Best,

Dak

Summary for this week: do it now, do it fast, do it cheap

So, what does all this mean?

Prototypes are your friend. Keep it simple.

1) In developing any product, especially software, DO IT NOW. So pick the things you can do NOW and get them done. Not partially done, not sloppily done…DONE, as in IT WORKS! This frees you to do something more, different…if you keep slogging on the same old thing, you are trapped in diminishing returns.

2) Decide when you are ready to show something GOOD to your customer. Keep the pace up. Don’t fall into the trap of “We can’t ship it, because the next version will be so much better!”. Show it, agree on what has to happen next[*], and then ship when you say you will.

3) Ship it, and follow up. Make a list of things you can do NOW. See step 1.

[*] Agreement means “We will pay you money when it does this.” Not “It would be nice if it did this”…that’s not agreement, that’s just socializing. You need agreement, or you aren’t DONE.

How does this fit into what I was talking about before?

A really good prototype might well be something you can sell; if not you can sell the idea, get feedback, and make something that can be taken to the market. It also allows you to play with something completely new, or the crazy idea for your main product that is too risky to actually DO with the main product.

As always, comments are more than welcome.

Best,
Dak

Dan Keldsen: Innovate or die

My brother, who is a Director with AIIM, presented this at the AIIM conference in Boston this week. I thought I’d share it here, as it directly relates to how costs inflect the innovation/return curve.Thanks,
Dak

Do Or Die Innovation By Process Based Information Management

From: dan.keldsen, 2 days ago

| View | Upload your own

Another “hyper-keynote” – although about half the the slides I’ve been using lately. 75 slides in 50 minutes, when done live. If you’re at the AIIM Conference in Boston this week (March 3rd, 2008), I’m giving this presentation on Tuesday, March 4th, at 2:30pm. Hope to see you there – and if note, feel free to comment here.

SlideShare Link