This makes me cringe:

If this error turns out to be an actual mistake Reinhart-Rogoff made, well, all I can hope is that future historians note that one of the core empirical points providing the intellectual foundation for the global move to austerity in the early 2010s was based on someone accidentally not updating a row formula in Excel.

I hear these reports of Excel errors [1] all the time but rarely do I hear of one that literally impacts the underpinnings for economic decisions across the entire world. In defense it sounds like this error by itself doesn’t make a huge difference but there are some other concerns with the model that, if true, means it is inconclusive at best. Public service announcement: always triple check your formulas.

The second thing I want to point out is that Excel is still the most indispensable piece of software in almost every business and government organization. The humble spreadsheet, nearly forty years old, runs the world’s economy. Considering this academic study shows that almost 90% of spreadsheets carry errors and, well, that should make us all want to hide our money in a box buried in the backyard.

Finally, many a word has been written over the past months regarding Microsoft’s “no compromise” Windows 8. Most have rightly pointed out that making no compromises is a compromise in and of itself and that legacy mode is horrible. But Excel specifically (and Office in general) is exactly why Microsoft made this decision. Could Microsoft have given users some of Excel in the Metro interface? Absolutely. But some of Excel isn’t good enough. When it comes to the power users who spend lots of money with Microsoft, Excel is an all or nothing proposition.

—

[1] I’m referring to data errors, not errors of Excel’s calculation engine. Excel does, however, have some calculation issues mostly because of backwards compatibility. As one example, when Visicalc wrote the original Net Present Value it actually didn’t calculate NPV; it calculated discounted cash flows. Lotus 1-2-3 copied it and so did Excel. =NPV() in Excel needs a modification to correctly calculate Net Present Value. (A great explanation is here.)

In February my post Apple’s Churning of the Gut was linked to by John Gruber at Daring Fireball and briefly went to the front page of Hacker News [1]. This is the third time since I started the blog that John has linked to one of my articles. It is an honor each time.

I’ve been curious, though, regarding the long-term impact of these kinds of links. The first two times I was writing only once a week. Since at the time of this third link I was writing every day, I was curious if those who clicked through would stick around. The short answer: no. See this graph [2]:

There is absolutely zero up-tick in the numbers for my own blog in the months following the Daring Fireball post.

Honestly, I’m surprised. I found a lot of interesting people to follow once I started reading John’s blog, especially in the early days of iOS development when John linked to far more independent developers [3]. I thought a few would have stuck around but when looking at the numbers, there is literally zero change in readership.

One possible explanation is that these users are RSS subscribers. I don’t believe WordPress is counting RSS subscribers correctly (if at all) in these numbers and therefore the count is not accurate. Given that I don’t truncate my RSS feed [4], they very well may not be counted. (Update: I found my Google Reader following, which is 3x my daily web view rate. Others report Google Reader accounts for 1/2 their RSS following. I have no idea what this number was before the Daring Fireball link.)

None of this is to diminish from the absolute rush of having one of my posts recognized by what I consider the leading publication on Apple’s perspective. I will admit it is a huge time-sink for a day or two, though, as I combed through more comments then this blog has had in its lifetime! Given that, John, I’m happy to make the sacrifice any time. 🙂

—

[1] This was a funny experience, by the way. The Hacker News community pretty much concluded I was a fanboy idiot while Daring Fireball’s community was in total agreement with my conclusions. Thus goes the holy wars of mobile operating systems.

[2] I believe the light blue is page views while the dark blue is unique visitors.

[3] John seems to link most often to a select group of publications and friends. This is completely gut feel. He may actually link to a wider group than I give credit. I may just be more familiar with the names now.

[4] If you are unaware, you can set  the RSS feed to only include a portion of the post. This means the user has to come to the site, via a “Read More” link, to read the rest of the post. In addition, Google Reader, the leading RSS feed reader (for now since Google announced its cancellation), only reads the post once and then makes it available to everyone who has subscribed to it from that one read. Thus everyone who follows me on Google Reader may very well only count as one view.