Mike Schaeffer's Blog

Microsoft has just announced a cool new feature on the Excel 12 blog: the databar. I think a picture can explain it better than I can:

Note from August 2025: This picture was formerly linked from Microsoft's Excel 12 blog, which borrowed it from another site. All that's gone now, but thanks to the magic of the Wayback Machine, I've found a copy to host locally.

This will be a nice way to look for trends/outliers, but I can also see it being useful for tracking parallel completion percentages in status reports, etc. Of the Excel 12 features announced so far, this is the one that I'm the most excited about. Of course, it's also the one that's easiest to approximate in Excel <12. Andrew has an approach using Autoshapes on his blog, and I'm going to present a slightly different approach.

IMO, his approach looks a lot better, this approach has the benefit of updating automatically. Pick your poison. It all centers around this little UDF:

Option Explicit

Function GraphBar(x As Double, _
                  Low As Double, _
                  High As Double, _
                  ScaleTo As Double) As String

    x = ((x - Low) / (High - Low)) * ScaleTo

    Dim i As Integer

    Dim blockFull As String
    Dim blockHalf As String

    blockFull = ChrW(9608)
    blockHalf = ChrW(9612)

    GraphBar = ""

    For i = 1 To Fix(x)
        GraphBar = GraphBar + blockFull
    Next

    If x - Fix(x) > 0.5 Then
        GraphBar = GraphBar + blockHalf
    End If
End Function

This isn't rocket science: all it does is rescale x from the range [Low, High] to the range [0.0, ScaleTo]. Then, it strings together that many Chrw(9608)'s, followed by a Chrw(9612), if the scaled value's fractional part is >0.5. The trick in this is that Chrw(9608) and Chrw(9612) are VBA expressions that produce the the Unicode equivalent of the old line drawing characters IBM put in the original PC [1]. 9608 is a full box ("█"), 9612 is a half box on the left ("▌"). The result of this function ends up being a string that (when displayed as Arial) looks like a horizontal bar. ("████▌"). Put a few of those in adjacent cells, and you get this:

The formula in C2 (and filled down) is =GraphBar(B2,MIN(B$2:B$8),MAX(B$2:B$8),5). The MIN and MAX set the scale, the 5 sets the maximum length of a bar. The maximum length, font size, column width can be tweaked to produce a reasonably attractive result, although I do reccomend using vertical centering.

If you want to get a little fancier, conditional formatting works on plot cells...

...whitespace can possibly improve the appearance...

...and this technique can scale.

1] (The original PC didn't have standard graphics, it was an option. If you bought the monochrome, non-graphics, video board, characters like this were as close as you could get to a bar chart.)

I've found a couple interesting websites related to computer history. The first is Dusty Decks, a blog related to some efforts to reconstruct Lisp and FORTRAN history. A highlight of this is a discussion on the Birth of the FORTRAN subroutine. Also via Dusty Decks is a website on the early history of the Lisp Programming Language.

That leads me to a couple books I've been reading lately. The first is Lisp in Small Pieces, by Christian Queinnec. I'm only a couple chapters in (stuck on continuations right now), but it's already been pretty profound. So far, the aspect of the book that's been the most useful is that it has gone through several core design choices Lisp implementors have to make ( Lisp-1 vs. Lisp-2, Lexical Scope vs. Dynamic Scope, types of continuations to support), and goes into depth regarding the implications and history of the choices involved. I think I'm finally starting to understand more of the significance of funcall and function in Common Lisp, not to mention throw/catch and block/return-from.

The second book is is The First Computers–History and Architectures, edited by Raul Rojas. This book is a collection of papers discussing the architecture of significant early computers from the late 30's and 40's. The thing that's so unique about the book is that it focuses on the architectural issues surrounding these machines: the kinds of hardware they were built with, how they processed information, and how they were programmed. Just as an example, it has a detailed description of many of ENIAC's functional units, even going into descriptions of how problems were set up on the machine. Another highlight of the book for me (so far) has been a description of Konrad Zuse's relay-based Z3, down to the level of a system architectural diagram, schematics of a few key circuits, and coverage of its microprogramming (!).

Group 1127, the group at Bell Labs that originally developed Unix, has been disbanded in a reorganization. I'm not exactly sure why this matters since all the original staff are gone and the remnants of systems research are elsewhere, but it did make the front page of Slashdot.

For some reason, I've been thinking a lot lately about Seymour Cray. When I was growing up, I remember asking my dad about who made the fastest computers in the world, and the answer at the time was Cray. I don't know if he meant the man or the company, but for a while both were true. I suppose it made an impression.

I've found a bunch of good things online about the man and his work:

• http://research.microsoft.com/users/gbell/craytalk/
• http://en.wikipedia.org/wiki/Seymour_Cray
• http://www.cwheroes.org/oralhistoryarchive/seymour_cray/index.asp
• http://www.spikynorman.dsl.pipex.com/CrayWWWStuff/

Reading through them, a couple of things made impressions on me:

• He didn't mind throwing bad ideas away (or saving them for later). The Cray 1 took a very different approach from the CDC 8600.
• Cray failed a lot. He was always pushing the limits and taking risks, and paid the price of those risks. The CDC 8600 failed, as did several designs for the Cray 2. The Cray 3 failed to sell, and the 4 doesn't seem to have hit the prototype stage at all. Even the Cray 2 doesn't seem to have been an unqualified success, thanks to issues with memory bandwidth.
• He had a very 'startup mentality'. His career seems to be a repeating story of initial success, spin off lab, and spin off company.
• A lot of his design problems weren't electronic at all. He seems to have struggled as much (if not more) with packaging and cooling as with anything else.
• He had a keen sense of style. With the possible exception of the Connection Machine CM-1/2, his machines were the most visually striking of the major supercomputers. Maybe it's superficial, but it can't have hurt the sales or publicity.
• He knew what he had accomplished. There's a story about his suprise when Steve Chen developed the X-MP from the Cray 1 and doubled (?) the performance. Of course, the story goes on to describe how Cray ended up appreciating the new design.

Anyway, I have nothing but the utmost respect for the man and his accomplishments. R.I.P, Mr. Cray.