Prior posts in this series:
Having discussed the separation of the tasks of writing a book and formatting the content, and the tools I use for writing, I now turn to the more complicated issue: formatting a book in a professional manner. Essentially, this is the realm of typography, which Wikipedia defines as: the art and technique of arranging type to make written language legible, readable, and appealing when displayed. The basic goal can be summarised by a simple principle: typography should never distract from the text; or, to put it slightly differently, unless he is specifically analysing it, the reader should never notice the typography. Wikipedia says essentially the same thing: Traditionally, text is composed to create a readable, coherent,
and visually satisfying typeface that works invisibly, without the
awareness of the reader. Even distribution of typeset material, with a
minimum of distractions and anomalies, is aimed at producing clarity and
transparency.
Typography should make the words and the sense of the text easy to discern: every time that the reader is forced to slow the movement of his ocular scan of a page -- or, even worse, if he is force to go back and re-read some text simply because of its layout -- the typography has failed.
(Some typography is simply inherently appalling: I stopped reading commercial e-books when, as I was reading a thriller from St. Martin's Press, I came across the word "you" -- typeset as "y-" on one page and "ou" on the next. It's hard to believe that such an insult to good taste was permitted by a supposedly competent commercial publishing house. How can anyone be expected not to give up in disgust when faced with such gratuitously atrocious typography? But I digress.)
Unsurprisingly, the single most important typographic decision is that of which typeface to use. As described in the colophon of most of my books, I generally use fonts from the Latin Modern family, which I find to be particularly beautiful and easy to read. Many people use fonts based on Times New Roman, but such fonts, designed as they were for the narrow columnar format of a newspaper, tend to lead to some confusion when used to print items such as novels. For example, the letter pair rn, which appears rather frequently in English, may be easily confused with the single letter m unless the reader is paying more attention than is comfortable when reading a fast-paced novel.
Almost all modern books are printed in a serif typeface. Books printed in sans serif typefaces are tiring and frustrating to read, and it is a mystery to me why any competent publishing house would ever print a book in such a way. (My best guess is that on the few occasions that it happens, it is because an enPetered editor with too much power wants to appear modern and gives no thought to the unnecessary difficulties he or she is foisting on the book's readers.) If you doubt the increase in blood pressure induced by reading a book printed in such a way, I suggest that you read Marcus Trescothick's otherwise wonderful autobiography about his life as quite possibly the best [cricket] batsman in the world as he battled with depression, Coming Back to Me. Some editor at HarperCollins decided to allow the book to be typeset in a sans serif font. That person cannot possibly have sat down and tried to read the resulting book.
Here is a sample page from Coming Back to Me (although this is far from doing justice to the experience of reading a page in the physical book):
I find that simply reading a page of this is slow work, probably because the shapes of the letters are so unusual that each word requires an abnormal amount of work for the brain to decode. After a few pages of this, I also find myself experiencing odd optical effects, such as a waviness in the baselines.
It's not just commercial houses that have indulged in foisting such poorly-considered typography on customers; even such a formerly august publisher as the OUP has done so. The OUP publishers the oddly uneven Very Short Introduction series, and some of those books (strangely, in this series there is no consistent layout from book to book) are typeset with a sans serif typeface. Here is a sample from A Very Short Introduction to Logic:
(The layout of this page violates a number of long-standing typographical conventions, leading me to suspect that the reader is again being subjected to an editor who believes in being modern rather than effective.)
Digressing wildly, the OUP has the distinction of publishing what is, in my opinion, the worst book I have read in recent years: An Introduction to Quantum Computing. Rife with grammatical and factual errors, it is hard to believe that any editor ever did more than simply skim the content without bothering to try to understand it. In my mind the final straw is the figures, of which this page contains two typical examples:
Quantum circuits are drawn throughout the book with the lines that indicate connections displayed in such a light colour that they can barely be seen, even when one knows where they must be. (But , strangely, not all circuits are drawn that way; every now and then one comes across one that is actually legible at a glance.) Can you see the boxes surrounding the gates and the connections between them without peering carefully at Figure 5.3 above? And notice that the text mentions "dashed boxes in Figure 5.5". Here is the page containing that figure:
I'll have to take the authors' word for it that there are dashed boxes in Figure 5.5, as I can't see them. None of this is a trick of the reproduction process for making the pages available here -- in fact, here one can magnify the images and see detail that is effectively invisible to the naked eye on the original page. In the original, the lines are printed too thinly in a very light grey. Given that such quantum circuits are in a real sense the essence of the book, it is bizarre that no care was taken to ensure that they are clear and unambiguous. The conclusion I draw is that no one, simply no one -- none of the three authors, nor any editor -- bothered even to sample the book when it was in the galley stage. Or perhaps no one subscribed to the notion that figures are supposed to demonstrate ideas clearly, rather than leave the reader perplexed because important features are essentially invisible. At least there is some consistency here, since the descriptions of important concepts in the text too frequently include substantive errors, or are presented in language that is so full of grammatical errors as to leave at least this reader puzzled as to the details of what the authors are trying to convey.
I have strayed too far from the intended subject, so will end there. Next time we'll begin to look at the more subtle aspects of good typography, which all too often are ignored nowadays, to the reader's detriment.
2017-04-24
2017-04-21
Call Busts and Reverse Busts in CQ WW, 2011
This is the seventh in a series of posts on busts and reverse busts in the CQ WW contests. These posts are based on the augmented versions of the CQ WW public logs.
Prior posts in the series:
In tables of reverse busts, one sometimes finds what seems like an unreasonable number of reverse busts (as, in this table, for UN7E/7). This is generally caused by a discrepancy between the call actually sent by the listed station and the one recorded as being sent in at least some QSOs in the station's log.
Prior posts in the series:
| 2011 SSB -- Most Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Busts | % Busts |
| 1 | EB1WW | 10,210 | 222 | 2.2 |
| 2 | LY7A | 9,729 | 154 | 1.6 |
| 3 | PJ2T | 13,021 | 147 | 1.1 |
| 4 | HG1S | 8,226 | 146 | 1.8 |
| 5 | OT5A | 12,310 | 145 | 1.2 |
| 6 | C5A | 18,155 | 141 | 0.8 |
| 7 | C37N | 15,766 | 139 | 0.9 |
| 8 | DP6T | 8,332 | 137 | 1.6 |
| 9 | EF8R | 9,718 | 136 | 1.4 |
| 10 | 9A1P | 14,403 | 136 | 0.9 |
| 2011 SSB -- Most Reverse Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Reverse Busts | % Reverse Busts |
| 1 | DF0HQ | 11,100 | 287 | 2.6 |
| 2 | JA5FDJ | 9,831 | 218 | 2.2 |
| 3 | UN7E/7 | 254 | 214 | 84.3 |
| 4 | SN3R | 6,724 | 108 | 3.1 |
| 5 | EE9Z | 9,998 | 173 | 1.7 |
| 6 | WE3C | 8,675 | 171 | 2.0 |
| 7 | C37N | 15,766 | 167 | 1.1 |
| 8 | HG1S | 8,226 | 165 | 2.0 |
| 9 | WW2Y | 1,459 | 160 | 11.0 |
| 10 | W2PV | 7,507 | 146 | 1.9 |
In tables of reverse busts, one sometimes finds what seems like an unreasonable number of reverse busts (as, in this table, for UN7E/7). This is generally caused by a discrepancy between the call actually sent by the listed station and the one recorded as being sent in at least some QSOs in the station's log.
| 2011 SSB -- Highest Percentage of Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Busts |
| 1 | RK0SM | 306 | 14.7 |
| 2 | RK9FBO | 160 | 14.4 |
| 3 | K2JMY | 188 | 13.3 |
| 4 | SP9IBJ | 307 | 12.1 |
| 5 | PV8ADI | 427 | 11.9 |
| 6 | LA9QNA | 110 | 11.8 |
| 7 | LU6EBY | 138 | 11.6 |
| 8 | K8TS | 183 | 11.5 |
| 9 | W7EWB | 117 | 11.1 |
| 10 | N5VEZ | 190 | 11.1 |
| 2011 SSB -- Highest Percentage of Reverse Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Reverse Busts |
| 1 | UN7E/7 | 254 | 84.3 |
| 2 | KB2HSH | 100 | 13.0 |
| 3 | WW2Y | 1,459 | 11.0 |
| 4 | BD4GNV | 118 | 10.2 |
| 5 | SM0LIU | 183 | 9.8 |
| 6 | PU2KXM | 147 | 9.5 |
| 7 | CE3TKV | 101 | 8.9 |
| 8 | BD3MF | 237 | 8.9 |
| 9 | 2W0BRR | 302 | 8.3 |
| 10 | OZ8KEL | 112 | 8.0 |
| 2011 CW -- Most Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Busts | % Busts |
| 1 | PV8ADI | 2,125 | 248 | 11.7 |
| 2 | OZ5E | 8,386 | 204 | 2.4 |
| 3 | EF8R | 8,612 | 181 | 2.1 |
| 4 | ZF1A | 7,846 | 170 | 2.2 |
| 5 | PJ2T | 15,684 | 160 | 1.0 |
| 6 | C5A | 19,616 | 153 | 0.8 |
| 7 | PI4DX | 7,205 | 151 | 2.1 |
| 8 | YE2S | 3,072 | 150 | 4.9 |
| 9 | HG1S | 9,211 | 148 | 1.6 |
| 10 | 6V7V | 4,000 | 143 | 3.6 |
| 2011 CW -- Most Reverse Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Reverse Busts | % Reverse Busts |
| 1 | DF0HQ | 12,780 | 391 | 3.1 |
| 2 | IR4X | 9,039 | 346 | 3.8 |
| 3 | JS3CTQ | 3,486 | 329 | 9.4 |
| 4 | ED9M | 8,355 | 311 | 3.7 |
| 5 | DR1A | 14,457 | 299 | 2.1 |
| 6 | TO7A | 7,044 | 239 | 3.4 |
| 7 | PW7T | 10,329 | 133 | 2.3 |
| 8 | P40F | 7,513 | 186 | 2.5 |
| 9 | HG1S | 9,211 | 182 | 2.0 |
| 10 | C6AAW | 13,483 | 179 | 1.3 |
| 2011 CW -- Highest Percentage of Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Busts |
| 1 | OH1FJ | 159 | 35.8 |
| 2 | W2GN | 319 | 23.5 |
| 3 | OM4DA | 400 | 21.0 |
| 4 | N9GUN | 111 | 20.7 |
| 5 | SM3AF | 105 | 20.0 |
| 6 | US7ZM | 136 | 19.1 |
| 7 | N8WS | 160 | 18.1 |
| 8 | PU1MMZ | 249 | 18.1 |
| 9 | EA2NA | 306 | 18.0 |
| 10 | JA3AHY | 123 | 17.9 |
| 2011 CW -- Highest Percentage of Reverse Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Reverse Busts |
| 1 | IW0EYT | 132 | 14.4 |
| 2 | LZ1195IR | 726 | 11.6 |
| 3 | GI0RQK | 1,506 | 9.6 |
| 4 | JS3CTQ | 3,486 | 9.4 |
| 5 | F8CRH | 1,698 | 9.4 |
| 6 | IK1TWC | 138 | 9.4 |
| 7 | JA2HYD | 133 | 9.0 |
| 8 | SA6G | 1,116 | 8.6 |
| 9 | DJ1SL | 201 | 8.5 |
| 10 | IZ2OBS | 142 | 8.5 |
2017-04-19
Revised Versions of the LANL GPS Charged-Particle Dataset
In the most recent post on this subject, I described various versions of the LANL GPS charged-particle dataset designed to improve the quality of the dataset as compared to the quality of the original release. Here, I intended to continue that process, but, as I describe below, due to some invalid data in the original dataset, I have reprocessed the data and uploaded a new file containing the resultant records.
First, though, I should mention that I received confirmation via private communication with LANL that, as I suspected, the value of collection_interval is intended to be taken from the set { 24, 120, 240, 4608 }, thus supporting the processing performed last time for stage 4.
However, I also received notification that in the original file ns41_041226_v1.03.ascii, the day number is incorrect for all the data marked as being in 2005, except for the very first entry. That is, everything past line 2330 of the original file is invalid.
The easiest way to remove the bad data seems to be (unfortunately) to go back to the the beginning of the processing and simply remove all the lines past line 2330 in file ns41_041226_v1.03.ascii, since I can think of no easy, foolproof way to remove the erroneous lines from the files created in stage 4 (or, indeed, any earlier stage).
Accordingly, we'll go back to the beginning and create new stages as follows:
sed -i '2331,99999d' -i ns41_041226_v1.03.ascii
We then execute several stages as before (except that what was stage n now becomes stage n+1) -- see the last post on this subject for details.
The next two stages are no longer necessary, as removing the tail of file ns41_041226_v1.03.ascii removed the records with bad values of decimal_day or collection_interval from the dataset. For the sake of consistency, though, I have retained the stages, although if you are duplicating the dataset there should be no need to create them (simply use the stage 3 files instead):
Here are the number of records for each satellite at the end of the processing for stage 5:
As a checkpoint, I have uploaded the new stage 5 dataset (I have also deleted the old stage 4 dataset). The MD5 checksum of this file is 774cc449fbd6fac836e17196cdaac363.
First, though, I should mention that I received confirmation via private communication with LANL that, as I suspected, the value of collection_interval is intended to be taken from the set { 24, 120, 240, 4608 }, thus supporting the processing performed last time for stage 4.
However, I also received notification that in the original file ns41_041226_v1.03.ascii, the day number is incorrect for all the data marked as being in 2005, except for the very first entry. That is, everything past line 2330 of the original file is invalid.
The easiest way to remove the bad data seems to be (unfortunately) to go back to the the beginning of the processing and simply remove all the lines past line 2330 in file ns41_041226_v1.03.ascii, since I can think of no easy, foolproof way to remove the erroneous lines from the files created in stage 4 (or, indeed, any earlier stage).
Accordingly, we'll go back to the beginning and create new stages as follows:
Stage 1: Remove all lines past line 2330 in the file ns41_041226_v1.03.ascii
This is easily accomplished simply by transferring all the data files to the stage-1 directory unchanged, and then applying the command:sed -i '2331,99999d' -i ns41_041226_v1.03.ascii
We then execute several stages as before (except that what was stage n now becomes stage n+1) -- see the last post on this subject for details.
Stage 2: All the data for each satellite in a single file
Stage 3: Remove all records marked as bad
The next two stages are no longer necessary, as removing the tail of file ns41_041226_v1.03.ascii removed the records with bad values of decimal_day or collection_interval from the dataset. For the sake of consistency, though, I have retained the stages, although if you are duplicating the dataset there should be no need to create them (simply use the stage 3 files instead):
Stage 4: Remove all records marked with invalid day of year
Stage 5: Correct time information
Here are the number of records for each satellite at the end of the processing for stage 5:
| Satellite | Stage 5 Records |
|---|---|
| ns41 | 1,990,891 |
| ns48 | 1,105,991 |
| ns53 | 1,331,687 |
| ns54 | 1,939,151 |
| ns55 | 1,055,328 |
| ns56 | 1,680,887 |
| ns57 | 1,082,626 |
| ns58 | 1,175,519 |
| ns59 | 1,516,528 |
| ns60 | 1,495,541 |
| ns61 | 1,470,445 |
| ns62 | 775,535 |
| ns63 | 652,110 |
| ns64 | 344,702 |
| ns65 | 480,935 |
| ns66 | 446,801 |
| ns67 | 327,994 |
| ns68 | 306,513 |
| ns69 | 262,992 |
| ns70 | 110,971 |
| ns71 | 221,336 |
| ns72 | 182,332 |
| ns73 | 145,858 |
As a checkpoint, I have uploaded the new stage 5 dataset (I have also deleted the old stage 4 dataset). The MD5 checksum of this file is 774cc449fbd6fac836e17196cdaac363.
2017-04-17
Call Busts and Reverse Busts in CQ WW, 2010
This is the sixth in a series of posts on busts and reverse busts in the CQ WW contests. These posts are based on the augmented versions of the CQ WW public logs.
Prior posts in the series:
In tables of reverse busts, one sometimes finds what seems like an unreasonable number of reverse busts (as, in this table, for ON46CQ and W3N0 [sic]). This is generally caused by a discrepancy between the call actually sent by the listed station and the one recorded as being sent in at least some QSOs in the station's log.
Prior posts in the series:
| 2010 SSB -- Most Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Busts | % Busts |
| 1 | EF8R | 12,741 | 170 | 1.3 |
| 2 | A73A | 10,884 | 151 | 1.4 |
| 3 | HG1S | 7,318 | 151 | 2.1 |
| 4 | OT5A | 8,687 | 149 | 1.7 |
| 5 | TM1O | 5,442 | 148 | 2.7 |
| 6 | JA7YRR | 4,822 | 144 | 3.0 |
| 7 | LZ9W | 10,350 | 126 | 1.2 |
| 8 | PJ2T | 14,892 | 125 | 0.8 |
| 9 | GM0B | 7,273 | 118 | 1.6 |
| 10 | PW7T | 9,182 | 110 | 1.2 |
| 2010 SSB -- Most Reverse Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Reverse Busts | % Reverse Busts |
| 1 | ON46CQ | 566 | 463 | 81.8 |
| 2 | T70A | 8,424 | 302 | 3.6 |
| 3 | W3N0 | 251 | 231 | 92.0 |
| 4 | UW2I | 2,691 | 171 | 6.4 |
| 5 | DF0HQ | 11,290 | 169 | 1.5 |
| 6 | JA3YBK | 5,291 | 160 | 3.0 |
| 7 | HG10P | 3,850 | 154 | 4.0 |
| 8 | SJ2W | 5,002 | 150 | 3.0 |
| 9 | RZ3AXX | 4,193 | 144 | 3.4 |
| 10 | S52ZW | 5,015 | 118 | 2.4 |
In tables of reverse busts, one sometimes finds what seems like an unreasonable number of reverse busts (as, in this table, for ON46CQ and W3N0 [sic]). This is generally caused by a discrepancy between the call actually sent by the listed station and the one recorded as being sent in at least some QSOs in the station's log.
| 2010 SSB -- Highest Percentage of Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Busts |
| 1 | PU2TRX | 123 | 14.6 |
| 2 | EA7IPP | 273 | 13.9 |
| 3 | LZ1VVV | 168 | 11.9 |
| 4 | OM8KD | 173 | 11.6 |
| 5 | BD7KQT | 231 | 11.3 |
| 6 | OH1FJ | 178 | 11.2 |
| 7 | IK5ZQC | 108 | 11.1 |
| 8 | BH4RIC | 110 | 10.9 |
| 9 | E76EA | 115 | 10.4 |
| 10 | YO7LTQ | 204 | 10.3 |
| 2010 SSB -- Highest Percentage of Reverse Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Reverse Busts |
| 1 | W3N0 | 251 | 92.0 |
| 2 | ON46CQ | 566 | 81.8 |
| 3 | OM4JD | 100 | 56.0 |
| 4 | BD1EFO | 125 | 16.0 |
| 5 | KH6SP/W5 | 532 | 13.0 |
| 6 | OK2SWD | 133 | 11.3 |
| 7 | UU1AZ | 330 | 11.2 |
| 8 | DX33CA | 108 | 10.2 |
| 9 | BA6QR | 118 | 10.2 |
| 10 | IW2MYH | 131 | 9.9 |
| 2010 CW -- Most Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Busts | % Busts |
| 1 | RC3W | 3,910 | 163 | 4.2 |
| 2 | M2A | 2,399 | 139 | 5.8 |
| 3 | HD2M | 7,858 | 135 | 1.7 |
| 4 | LY7A | 7,750 | 129 | 1.7 |
| 5 | 9A1A | 10,869 | 129 | 1.2 |
| 6 | AN2A | 4,866 | 122 | 2.5 |
| 7 | PW7T | 7,943 | 115 | 1.4 |
| 8 | KP2M | 7,722 | 112 | 1.5 |
| 9 | OZ5E | 6,358 | 110 | 1.7 |
| 10 | C4N | 7,138 | 108 | 1.5 |
| 2010 CW -- Most Reverse Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Reverse Busts | % Reverse Busts |
| 1 | RZ3VO | 1,020 | 892 | 87.5 |
| 2 | ED9M | 7,829 | 316 | 4.0 |
| 3 | DF0HQ | 11,035 | 283 | 2.6 |
| 4 | GI0RQK | 1,598 | 283 | 17.7 |
| 5 | JS3CTQ | 2,419 | 258 | 10.7 |
| 6 | ES9C | 7,188 | 213 | 3.0 |
| 7 | CN3A | 5,948 | 198 | 3.3 |
| 8 | IR4X | 7,928 | 184 | 2.3 |
| 9 | PW7T | 7,943 | 164 | 2.1 |
| 10 | IU1A | 3,843 | 160 | 4.2 |
| 2010 CW -- Highest Percentage of Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Busts |
| 1 | OH1FJ | 211 | 26.5 |
| 2 | YO7LYM | 195 | 24.1 |
| 3 | F5BTH | 155 | 20.6 |
| 4 | SP9IBJ | 280 | 18.2 |
| 5 | JO3AGQ | 124 | 16.9 |
| 6 | LA8TIA | 108 | 16.7 |
| 7 | JA6WW | 127 | 16.5 |
| 8 | KJ4AOM | 187 | 15.5 |
| 9 | OH2MO | 280 | 15.4 |
| 10 | GM0TTY | 115 | 14.8 |
| 2010 CW -- Highest Percentage of Reverse Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Reverse Busts |
| 1 | RZ3VO | 1,020 | 87.5 |
| 2 | DL7VRG | 148 | 24.3 |
| 3 | GI0RQK | 1,598 | 17.7 |
| 4 | SO9R | 159 | 15.1 |
| 5 | KS2G | 128 | 13.3 |
| 6 | 2E0BFJ | 105 | 11.4 |
| 7 | KD3TB | 161 | 11.2 |
| 8 | JS3CTQ | 2,419 | 10.7 |
| 9 | OE1BKA | 115 | 10.4 |
| 10 | BD8NBG | 342 | 10.2 |
2017-04-14
A Grid-Based Scatter Metric for the RBN
I recently posted about an attempt to find a reasonable measure of the scatter of the stations that post reports to the Reverse Beacon Network (RBN). In that post, I looked at a metric based on the direct measurement of distances between posting stations. In this post, I look at a quite different approach based simply on counting the occupancy of cells defined over the 2-sphere that closely models the surface of the Earth.
If we look at two geographic representations of the RBN as it was in 2009 and then again in 2016...
...what strikes me most is not anything to do (directly) with the distribution of distances between points, nor is it to do with the colours of the points, but rather the obvious facts that: (i) there are a couple of areas where relatively many points are clustered; (ii) the vast majority of the area has no points at all; (iii) some spreading and filling-in has occurred over the time-span covered by the two figures.
That being so, it seems plausible that a better metric of the spreading of the RBN could be obtained by dividing the terrestrial 2-sphere into cells and counting the cells that are occupied (regardless of the number of posting stations within, or the number of posts emanating from, each cell). Such a metric has the additional merit of being relatively simple to calculate compared to the approach that derived a metric from the distances between posting stations.
Ideally, one would define the cells so that they would have two properties: (i) they would all be the same size; and (ii) they would all be as compact as possible.
The second desideratum may not be obvious: however, consider the case in which all the cells are defined such that each cell comprises all the longitudes between pairs of latitude lines (while defining the latitude lines in such a way that the area of each cell is the same, of course). One could easily divide the 2-sphere in such a manner, but this would mean that, for example, the cells for all the latitudes that cover Europe would appear as being populated, which would make a mockery of any attempt to interpret the number of populated cells as any kind of metric for the distribution of RBN posting stations.
To meet both desired criteria simultaneously essentially requires that we solve our friend from last time, the Tammes problem. Indeed, unless we allow non-compact cells, the problem of covering the surface of the Earth with cells of equal size results in a non-trivial distribution of cells (see, for example, the HEALPix projection and the documentation for the HEALPix code).
To make reasonable progress, we take a slightly different approach and relax our criteria, so that the cells are merely both reasonably similar in size and reasonably compact (after all, what we really want is a system that allows us quickly to decide in which cell a given RBN station is located, with the condition that reasonably separated RBN stations will be placed in different cells, rather than a system that meets any precise mathematical requirement).
The simplest approach seems to be (assuming latitudinal symmetry) to move from a pole towards the equator in bands defined by parallels of latitude, and defining the (integral) number of cells around each band in such a way that the size of the cells varies by only a relatively small amount.
For simplicity, we will consider just one half of the Earth's surface (the southern hemisphere); the pertinent results for the entire terrestrial 2-sphere will follow immediately by symmetry.
It's easy to show that the area 1A2 of that portion of a sphere of radius r between two co-latitudes λ1 and λ2 is:
$$_1A_2 = 2\pi r^2 (cos \lambda_1 - cos \lambda_2)$$
So, suppose that we divide the hemisphere into bands of latitude 10°, and wish to construct a grid with a total of 50 cells (i.e., 100 cells cover the entire 2-sphere) aligned along these bands. The resultant best-fit cells can be summarised as:
A similar table for latitude bands of 15° is similarly summarised:
In these tables, we labelled the column that shows the number of cells in each latitude band with the subscript 100 rather than 50, because that is the number of cells over the entire 2-sphere.
Note that these are just two of many tables that could be created for various values of the widths of the latitude bands and the total number of cells desired. We can denote any particular table by using parameters, so that G(15, 100) denotes the second table above, defining a total of 100 grid-based cells over the 2-sphere, with 15° bands of latitude. G(15, 100) seems like a reasonable metric with which to begin.
The cells for any latitude band may, of course, start at any longitude, but it makes things easier if, for each latitude band, we place a boundary at the Greenwich meridian, and also require that cell boundaries lie exactly on a meridian that is an integral number of degrees.
This gives us the following table for the longitudes at which cell boundaries occur:
With these constraints, G(15, 100) defines 100 cells as shown on this map:
These look like reasonable cells; therefore we will proceed with G(15,100) as the basis for the grid-based scatter metric. The actual value of the scatter metric over some time-frame Δ will therefore be the number of cells in the above plot that contain at least one RBN poster within the time-frame.
I have not addressed the issue of the number of cells we should use. There is no obvious criterion for choosing a particular number, although the number should be sufficiently large that it reflects the fact that large parts of the Earth's surface have no RBN posting stations. On the other hand, the number should not be so large that essentially every RBN node is in a distinct cell. The 100 cells on the above map seem a reasonable number to me. It also naturally leads to an index whose value, at least in theory, runs from 0 to 100.
We can now easily compare the results of using G(15, 100) and the previously-defined scatter metric that was calculated directly from the positions of the posting stations.
Simply plotting the values of the two metrics as a function of time, we obtain:
(Pearson correlation coefficient is 0.78; rate of increase is about 145 per year.)
(Pearson correlation coefficient is 0.91; rate of increase is about 2.2 per year.)
Plotting the values of the metrics as functions of the number of posting stations, we obtain:
(Pearson correlation coefficient is 0.75; rate of increase is about 4.8 per poster.)
(Pearson correlation coefficient is 0.94; rate of increase is about 0.078 per poster.)
From these plots, I conclude that there is likely no point to looking at the third category of possible scatter metric (described as "[m]etrics based indirectly on a distance metric" in the prior post on this subject). [By this phrase, I intended to indicate the use of more complicated measures of the distribution of the distance-based measurements, rather than just the mean separation: for example, looking at higher moments of the distribution, or seeing if anything useful could be gained by looking at the fractal dimensionality.] The G(15,100) scatter metric is easy to calculate, it seems to correspond closely to the intuitive understanding of "scatter", and it appears to represent at least as good a metric as a directly-calculated metric based on mean separation of posting stations.
Thus, in future postings that require the use of a scatter metric, I will use the G(15, 100) metric unless there appears to be a good reason to do otherwise.
If we look at two geographic representations of the RBN as it was in 2009 and then again in 2016...
...what strikes me most is not anything to do (directly) with the distribution of distances between points, nor is it to do with the colours of the points, but rather the obvious facts that: (i) there are a couple of areas where relatively many points are clustered; (ii) the vast majority of the area has no points at all; (iii) some spreading and filling-in has occurred over the time-span covered by the two figures.
That being so, it seems plausible that a better metric of the spreading of the RBN could be obtained by dividing the terrestrial 2-sphere into cells and counting the cells that are occupied (regardless of the number of posting stations within, or the number of posts emanating from, each cell). Such a metric has the additional merit of being relatively simple to calculate compared to the approach that derived a metric from the distances between posting stations.
Defining the Cells
Ideally, one would define the cells so that they would have two properties: (i) they would all be the same size; and (ii) they would all be as compact as possible.
The second desideratum may not be obvious: however, consider the case in which all the cells are defined such that each cell comprises all the longitudes between pairs of latitude lines (while defining the latitude lines in such a way that the area of each cell is the same, of course). One could easily divide the 2-sphere in such a manner, but this would mean that, for example, the cells for all the latitudes that cover Europe would appear as being populated, which would make a mockery of any attempt to interpret the number of populated cells as any kind of metric for the distribution of RBN posting stations.
To meet both desired criteria simultaneously essentially requires that we solve our friend from last time, the Tammes problem. Indeed, unless we allow non-compact cells, the problem of covering the surface of the Earth with cells of equal size results in a non-trivial distribution of cells (see, for example, the HEALPix projection and the documentation for the HEALPix code).
To make reasonable progress, we take a slightly different approach and relax our criteria, so that the cells are merely both reasonably similar in size and reasonably compact (after all, what we really want is a system that allows us quickly to decide in which cell a given RBN station is located, with the condition that reasonably separated RBN stations will be placed in different cells, rather than a system that meets any precise mathematical requirement).
The simplest approach seems to be (assuming latitudinal symmetry) to move from a pole towards the equator in bands defined by parallels of latitude, and defining the (integral) number of cells around each band in such a way that the size of the cells varies by only a relatively small amount.
For simplicity, we will consider just one half of the Earth's surface (the southern hemisphere); the pertinent results for the entire terrestrial 2-sphere will follow immediately by symmetry.
It's easy to show that the area 1A2 of that portion of a sphere of radius r between two co-latitudes λ1 and λ2 is:
$$_1A_2 = 2\pi r^2 (cos \lambda_1 - cos \lambda_2)$$
So, suppose that we divide the hemisphere into bands of latitude 10°, and wish to construct a grid with a total of 50 cells (i.e., 100 cells cover the entire 2-sphere) aligned along these bands. The resultant best-fit cells can be summarised as:
| Cells in 10° bands | ||||
|---|---|---|---|---|
| λ1(°) | λ2(°) | 1A2 | N100 | |
| 0 | 10 | 0.015192247 | 1 | |
| 10 | 20 | 0.0451151 | 2 | |
| 20 | 30 | 0.736672 | 4 | |
| 30 | 40 | 0.099981 | 5 | |
| 40 | 50 | 0.1232568 | 6 | |
| 50 | 60 | 0.1427876 | 7 | |
| 60 | 70 | 0.1579799 | 8 | |
| 70 | 80 | 0.168372 | 8 | |
| 80 | 90 | 0.1736482 | 9 | |
A similar table for latitude bands of 15° is similarly summarised:
| Cells in 15° bands | ||||
|---|---|---|---|---|
| λ1(°) | λ2(°) | 1A2 | N100 | |
| 0 | 15 | 0.0340742 | 2 | |
| 15 | 30 | 0.0999004 | 5 | |
| 30 | 45 | 0.1589186 | 8 | |
| 45 | 60 | 0.2071068 | 10 | |
| 60 | 75 | 0.241181 | 12 | |
| 75 | 90 | 0.258819 | 13 | |
In these tables, we labelled the column that shows the number of cells in each latitude band with the subscript 100 rather than 50, because that is the number of cells over the entire 2-sphere.
Note that these are just two of many tables that could be created for various values of the widths of the latitude bands and the total number of cells desired. We can denote any particular table by using parameters, so that G(15, 100) denotes the second table above, defining a total of 100 grid-based cells over the 2-sphere, with 15° bands of latitude. G(15, 100) seems like a reasonable metric with which to begin.
The cells for any latitude band may, of course, start at any longitude, but it makes things easier if, for each latitude band, we place a boundary at the Greenwich meridian, and also require that cell boundaries lie exactly on a meridian that is an integral number of degrees.
This gives us the following table for the longitudes at which cell boundaries occur:
| Longitudes of Cell Boundaries | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N Cells | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 |
| 1 | 360 | ||||||||||||
| 2 | 180 | 360 | |||||||||||
| 3 | 120 | 240 | 360 | ||||||||||
| 4 | 90 | 180 | 270 | 360 | |||||||||
| 5 | 72 | 144 | 216 | 288 | 360 | ||||||||
| 6 | 60 | 120 | 180 | 240 | 300 | 360 | |||||||
| 7 | 51 | 103 | 154 | 206 | 257 | 309 | 360 | ||||||
| 8 | 45 | 90 | 135 | 180 | 225 | 270 | 315 | 360 | |||||
| 9 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | 360 | ||||
| 10 | 36 | 72 | 108 | 144 | 180 | 216 | 252 | 288 | 324 | 360 | |||
| 11 | 33 | 65 | 98 | 131 | 164 | 196 | 229 | 262 | 295 | 327 | 360 | ||
| 12 | 30 | 60 | 90 | 120 | 150 | 180 | 210 | 240 | 270 | 300 | 330 | 360 | |
| 13 | 28 | 55 | 83 | 111 | 138 | 166 | 194 | 222 | 249 | 277 | 304 | 332 | 360 |
With these constraints, G(15, 100) defines 100 cells as shown on this map:
These look like reasonable cells; therefore we will proceed with G(15,100) as the basis for the grid-based scatter metric. The actual value of the scatter metric over some time-frame Δ will therefore be the number of cells in the above plot that contain at least one RBN poster within the time-frame.
Digression - How Many Cells?
I have not addressed the issue of the number of cells we should use. There is no obvious criterion for choosing a particular number, although the number should be sufficiently large that it reflects the fact that large parts of the Earth's surface have no RBN posting stations. On the other hand, the number should not be so large that essentially every RBN node is in a distinct cell. The 100 cells on the above map seem a reasonable number to me. It also naturally leads to an index whose value, at least in theory, runs from 0 to 100.
Results
We can now easily compare the results of using G(15, 100) and the previously-defined scatter metric that was calculated directly from the positions of the posting stations.
Simply plotting the values of the two metrics as a function of time, we obtain:
(Pearson correlation coefficient is 0.78; rate of increase is about 145 per year.)
(Pearson correlation coefficient is 0.91; rate of increase is about 2.2 per year.)
Plotting the values of the metrics as functions of the number of posting stations, we obtain:
(Pearson correlation coefficient is 0.75; rate of increase is about 4.8 per poster.)
(Pearson correlation coefficient is 0.94; rate of increase is about 0.078 per poster.)
From these plots, I conclude that there is likely no point to looking at the third category of possible scatter metric (described as "[m]etrics based indirectly on a distance metric" in the prior post on this subject). [By this phrase, I intended to indicate the use of more complicated measures of the distribution of the distance-based measurements, rather than just the mean separation: for example, looking at higher moments of the distribution, or seeing if anything useful could be gained by looking at the fractal dimensionality.] The G(15,100) scatter metric is easy to calculate, it seems to correspond closely to the intuitive understanding of "scatter", and it appears to represent at least as good a metric as a directly-calculated metric based on mean separation of posting stations.
Thus, in future postings that require the use of a scatter metric, I will use the G(15, 100) metric unless there appears to be a good reason to do otherwise.
2017-04-12
Call Busts and Reverse Busts in CQ WW, 2009
This is the fifth in a series of posts on busts and reverse busts in the CQ WW contests. These posts are based on the augmented versions of the CQ WW public logs.
Prior posts in the series:
In tables of reverse busts, one sometimes finds what seems like an unreasonable number of reverse busts (as, in this table, for RRA3XCZ and W0QG). This is generally caused by a discrepancy between the call actually sent by the listed station and the one recorded as being sent in at least some QSOs in the log.
Prior posts in the series:
| 2009 SSB -- Most Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Busts | % Busts |
| 1 | AO8A | 11,468 | 139 | 1.2 |
| 2 | Z37M | 7,164 | 139 | 1.9 |
| 3 | OT5A | 9,618 | 134 | 1.4 |
| 4 | LS2D | 3,939 | 119 | 3.0 |
| 5 | EK8WA | 5,032 | 110 | 2.4 |
| 6 | 9A1P | 11,677 | 118 | 1.0 |
| 7 | EB1WW | 6,051 | 115 | 1.9 |
| 8 | PJ2T | 11,207 | 104 | 0.9 |
| 9 | AM3SSB | 6,000 | 102 | 1.7 |
| 10 | HG1S | 7,398 | 102 | 1.4 |
| 2009 SSB -- Most Reverse Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Reverse Busts | % Reverse Busts |
| 1 | DF0HQ | 12,372 | 178 | 1.4 |
| 2 | AO8A | 11,468 | 175 | 1.5 |
| 3 | S52ZW | 6,096 | 134 | 2.2 |
| 4 | HB0/HB9AON | 7,705 | 128 | 1.7 |
| 5 | JR5VHU | 5,769 | 125 | 2.2 |
| 6 | TM2S | 3,388 | 111 | 3.3 |
| 7 | SE0X | 2,020 | 108 | 5.3 |
| 8 | UX2IO | 3,555 | 108 | 3.0 |
| 9 | EE9Z | 9,385 | 107 | 1.1 |
| 10 | DQ4W | 5,677 | 101 | 1.8 |
| 2009 SSB -- Highest Percentage of Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Busts |
| 1 | SQ8MFB | 129 | 17.8 |
| 2 | OH1FJ | 126 | 15.9 |
| 3 | LS4DX | 123 | 13.8 |
| 4 | EW8BQ | 273 | 12.1 |
| 5 | EI4JZ | 175 | 11.4 |
| 6 | BD1HFP | 115 | 11.3 |
| 7 | NT4TS | 171 | 11.1 |
| 8 | UT7VR | 198 | 10.6 |
| 9 | OZ6EI | 128 | 10.2 |
| 10 | EA3HAN | 119 | 10.1 |
| 2009 SSB -- Highest Percentage of Reverse Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Reverse Busts |
| 1 | LT0D | 214 | 13.6 |
| 2 | VE7RSV/P | 132 | 11.4 |
| 3 | 4Z5TK | 101 | 10.9 |
| 4 | SM7TZK | 310 | 10.0 |
| 5 | BG4IKE | 101 | 9.9 |
| 6 | 5P0O | 152 | 9.9 |
| 7 | K4GOP | 115 | 9.6 |
| 8 | YC1BYX | 170 | 9.4 |
| 9 | F1SMV | 135 | 8.9 |
| 10 | LU8EOT | 187 | 8.6 |
| 2009 CW -- Most Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Busts | % Busts |
| 1 | OZ5E | 3,706 | 162 | 4.4 |
| 2 | HG1S | 6,127 | 124 | 2.0 |
| 3 | EE2W | 10,410 | 115 | 1.1 |
| 4 | ZM4T | 3,249 | 107 | 3.3 |
| 5 | 6Y1V | 10,486 | 101 | 1.0 |
| 6 | EA5CW | 6,180 | 101 | 1.6 |
| 7 | RC3W | 2,623 | 98 | 3.7 |
| 8 | YT2W | 6,760 | 97 | 1.4 |
| 9 | VQ5V | 4,660 | 96 | 2.1 |
| 10 | 4M5IR | 1,281 | 96 | 7.5 |
| 2009 CW -- Most Reverse Busts | ||||
|---|---|---|---|---|
| Position | Call | QSOs | Reverse Busts | % Reverse Busts |
| 1 | EE2W | 10,410 | 438 | 4.2 |
| 2 | W0QG | 420 | 352 | 83.8 |
| 3 | DF0HQ | 9,945 | 254 | 2.6 |
| 4 | RRA3XCZ | 277 | 252 | 91.0 |
| 5 | JS3CTQ | 1,926 | 220 | 11.4 |
| 6 | JS6RGY | 2,501 | 180 | 7.2 |
| 7 | DR1A | 10,163 | 166 | 1.6 |
| 8 | IR4X | 7,391 | 158 | 2.1 |
| 9 | CN3A | 4,559 | 154 | 3.4 |
| 10 | EF8M | 7,419 | 152 | 2.0 |
In tables of reverse busts, one sometimes finds what seems like an unreasonable number of reverse busts (as, in this table, for RRA3XCZ and W0QG). This is generally caused by a discrepancy between the call actually sent by the listed station and the one recorded as being sent in at least some QSOs in the log.
| 2009 CW -- Highest Percentage of Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Busts |
| 1 | CE6VMO | 129 | 20.9 |
| 2 | YO7LTQ | 133 | 19.5 |
| 3 | K9IA | 101 | 18.8 |
| 4 | N1IMW | 138 | 18.1 |
| 5 | N7TL | 127 | 16.5 |
| 6 | OH2MO | 146 | 16.4 |
| 7 | KB0R | 123 | 15.4 |
| 8 | W9IIX | 138 | 14.5 |
| 9 | PY3CAL | 152 | 14.5 |
| 10 | G4RYV | 170 | 14.1 |
| 2008 CW -- Highest Percentage of Reverse Busts (≥100 QSOs) | |||
|---|---|---|---|
| Position | Call | QSOs | % Reverse Busts |
| 1 | RRA3XCZ | 277 | 91.0 |
| 2 | JI30GI | 110 | 89.1 |
| 3 | W0QG | 420 | 83.8 |
| 4 | EI4JZ | 108 | 35.2 |
| 5 | DL7VRG | 172 | 22.1 |
| 6 | KS2G | 102 | 12.7 |
| 7 | JS3CTQ | 1,926 | 11.4 |
| 8 | PA3ANN | 290 | 10.3 |
| 9 | PY2ASS | 135 | 9.6 |
| 10 | UX3HA | 763 | 9.6 |
2017-04-10
Band-by-Band Comparison of DX QSO Rates for CQ WW CW and SSB, 2005 to 2016
The availability of public logs for CQ WW CW and SSB contests spanning twelve years allows us to look at the variation in the rate at which DX QSOs are made throughout each contest over a relatively long period. (I here define DX QSOs as those between different zones; other definitions are obviously possible.)
In the plots, the number of QSOs is accumulated every ten minutes, so there are six points per hour.
I think that the results are rather interesting. I'll let you draw your own conclusions.
In regard to the 10m results, I should point out that the CW contest occurs roughly a month after the SSB one, and is therefore considerably closer to the winter solstice, with the concomitant decrease in the number of hours of common daylight between Europe and North America (particularly western NA). I have long wondered what would happen if CQ WW (and CQ WPX and ARRL DX) would switch CW and SSB weekends every year.
In the plots, the number of QSOs is accumulated every ten minutes, so there are six points per hour.
I think that the results are rather interesting. I'll let you draw your own conclusions.
In regard to the 10m results, I should point out that the CW contest occurs roughly a month after the SSB one, and is therefore considerably closer to the winter solstice, with the concomitant decrease in the number of hours of common daylight between Europe and North America (particularly western NA). I have long wondered what would happen if CQ WW (and CQ WPX and ARRL DX) would switch CW and SSB weekends every year.
Subscribe to:
Posts (Atom)
















