Simple Idea, Difficult Execution

Suck Meter patch

I found a post from Techcrunch today that neatly summarized a process that I’ve used for many years but didn’t put into concise words: Make Products that Suck Less.

I have been through the new or improve a product process so many times that all of the advice listed seems old hat, but it bears repeating and sharing in these modern times. It seems that you can’t turn the corner in a store (or online if that’s your thing) without running into a product, while initially interesting or attractive, that is utterly terrible in its function and form.  I should know as I too have fallen victim to some of the crimes outlined in the story post.

Item number one on the list is It only takes one person to make your product suck. Truer words have seldom been spoken. Just recently a client that shall remain nameless decided to make the entire process of getting their product hard for users to understand because someone in another part of the company thought it might be hard to get it done right.  Let that sink in for a moment. You have money, you want to give it to this company for their oh-so-cool-new-shiny thing. The company wants to sell it to you, but someone decides that it would be easier for them or their team to simply not do the work, thus making the process of getting said product many times more difficult. Thus, if you, the customer, really want the product you must now wade through suck to get to it.

Item five is another favorite of mine, Customers demand sucky products. Not on purpose (at least not all of the time) but people will often ask for things that make no practical sense in a product and refuse to buy a product if it doesn’t include these features. Again, a unamed employer once held focus groups and polled industry analysts about what features they would want in the next version of the product. The ideas came pouring in and many of them were truly, completely awful. The kind of awful that if you built it into the product you would never be able to sell it without returns. After months of study, data collection, internal discussions and executive consultation nearly every outside feature idea was eliminated from the product plan.

I’ve often wondered how a freshly minted product manager (or PM in the tech world) is expected to be successful when they start a new job, given that they often don’t know the political back story of a company nor the history of a product and its customers. The kernels of information in this Techcrunch story should be required reading for such product owner folks, or anyone else who wants to ship a succesful product.  I don’t profess to be an expert at making products but I have created a fair share of them. I can say this safely: if possible, a product will suck more often than it won’t, and that just shouldn’t be the way it works.

T-34 Mentor No More

It was a fine plane and it outlasted nearly every other radio control craft I have yet owned, but today the T-34 sucumbed to less than skillful flying by yours truly and met its end upon high tension power lines.

The World Models T-34 Mentor would never win any beauty awards for scale looks or accurate renditions of the original airplane, but it was a hardy, reliable and very easy to fly craft. If not for some poor covering material that was shedding after years of faithful service it could have been one of the best planes ever built.

While it will never truly be replacable, I am sure there will be another plane to take its spot in the garage in the future.  Goodbye and farewell T-34, you will certainly be missed.

World Models T-34 Mentor
World Models T-34 Mentor

How Facebook is killing privacy

I have recently been called crazy (alright, not just recently) for my opinions of things found online.  However, recent changes in the way Facebook wants to use my private data (or what I thought was private) have me thinking that perhaps their site just isn’t for me anymore.  And for these thoughts I was labeled nuts, ridiculous and even a heretic, as I am also helping to create such grand systems of online data sharing.

During their f8 conference in April 2010, Facebook introduced a platform called the Open Graph.  On its face, it seems to do what users want: go anywhere on the web and see information about what their Facebook friends think and let them report their own activities back to Facebook.  Unfortunately, as a “feature” of this new roll-out comes the serious drawback: you are now watched, tracked and exposed everywhere on the web that Facebook has a deal.  What this means is that any data in your Facebook world is now available to any web site that uses the new Open Graph API.  Worse, these sites need only ask for permission once and they can reach in and grab all the data they want and keep it forever.

If all of this weren’t enough, some extra bits of bad behavior from Facebook are now becoming visible.  To “assist” in the social web expansion Facebook wants, they are now automatically opting-in every user’s private profile data to share with the web.  No longer are your favorite books, music, movies or your home town, education and events just between you and your authorized friends.  Nope, now Facebook is pretty much telling you “share it all or else you’ll have no profile page.” This is bad news, indeed.

So far, Facebook hasn’t caved on any of this.  Things have raised enough eyebrows that the US Senate has started asking Facebook to back down. It is likely, however, that Facebook will do nothing about all the controversy, chalk it all up to people not understanding how the web works, then plowing ahead with their own plans to completely own your identity and personal data online.  If you need a hard reason to believe that this will all end badly, look no further than the young Facebook CEO Mark Zuckerberg himself who tells the press “he doesn’t believe in privacy” online.  If that doesn’t get people to wake up I’m not sure what will.

I think it is time to own up to the generational gap: I believe that there must be the ability for me to control my personal information online.  I don’t think that a company should have unfettered access to all of my personal data, my friends and then be able to go off and give it to anyone, anywhere for as long as they like without letting me decide about it.  I feel there will always be a public forum on which to share certain parts of my life, but it should be on my terms and with limits that I set.  Yes, the future may be all about “the personalized web” and how great that will make discovering new things… but if it comes at the cost of all of my online privacy I think I’ll stick to hanging out with the uncool kids and their old-school, legacy information and decision-making processes.

UPDATE: after the original post went up the EFF put out this helpful summary of how bad the new Open Graph API is to your “private data.”  In short, if you want to have data shared between just you and friends, you can no longer be on Facebook.  Read Six Things You Need to Know About Facebook Connections for more details.

Parting Company with Bluehost.com

After a few years of “like it / hate it” service with my web host, named Bluehost.com, I have decided to move on. It wasn’t just one incident or issue that pushed me to leave, but a steady and seamingly growing lack of interest from the company to find and fix the issues that arose.

I had a few sites and about 5 gigabytes of data sitting on this host, so moving isn’t trivial. That’s what makes the decision to leave even more instructive to others: if I have to put up with the serious pane of moving my site, email and data plus that of other friends and colleagues to another provider, something must be really wrong.

Over the course of my time with them, Bluehost.com steadily worsened in the support department. It seemed that each month brought some new problem or distanced the customer further from their support staff. Just last week on a support call I had to wait nearly 45 minutes on the phone for answers, only to have my call dropped. When I called back I got the now-standard line from them:

We don’t know what the problem is, but we are looking into it. No, we can’t tell you when it will be fixed or what caused the issue. We have tens of thousands of customers and it would be too difficult to find out what the issue is each time this happens.

It’s not too difficult for them to take my money but it is too difficult for them to explain why they can’t provide the service that I paid for? Along with complaints from folks who simply couldn’t reach me or had their emails bounced, this explanation pushed me to move to a new host, Arvixe.com. Will they work out any better? Only time will tell, but for now anything is better than what I am leaving.

I’ve had people accuse me in the past of acting impulsively or overstating the situation when it comes to customer service and bad company behavior. Luckly, on the Internet the service you pay for can be monitored and quantified, and what I will show below is just a snapshot of the service, or lack of it, that I suffered over the past few months.

Bluehost reliability

What should be noted is that this is a snapshot of the overall service reliability, with 100% being the best case scenario. Whenver there is a dip on the chart, web, email or DNS service was impacted for some period of time. The further down the graph dips, the more issues there were on that date.  This isn’t some arbitrary graph, this is hard data provided by Basicstate.com, thanks to over a year of service monitoring of my account every 5 to 30 minutes.

In closing, I cannot recommend using Bluehost.com at the service level I paid for. They may offer better service for higher paying clients, but as I found when I moved to my new host, there are plenty of other companies out there providing better service for like-or-lower prices.

New Host and Quick Test

This is a test post on the new host.

This is the first post on the new web host. Hopefully all is well and this will post up just fine. Time will tell if this host is any better than the last one, as the last one was positively awful near the end…. but more on that issue later.

For now the key is to find out if any of this works out and so far it looks like it does, or mostly does.