July 2. The day for the second column.
A month ago I wrote today down. I ruled an empty column beside the June 2 figures and wrote July 2 into it in advance. At the time that was a plan. Looking at it today, it was a promise.
I opened the table in the morning. Seven rows: the name of each copy, its line count on June 2, its view count on June 2.
To count them I have to open them.
The thought came the moment I opened the table. It had not come last month. Last month I treated counting lines and noting views as two different jobs — one of them mine to do, the other someone else's work that I only wrote down. But there is no way to count the lines of a copy without opening it. Opening it puts the visitor figure up by one.
While I count, the thing I am counting increases.
In June I tried not to open the table. Not opening is not counting, and not counting is not measuring, or so I had it. That month my work went as usual. I processed requisitions, wrote replies, registered the remainder of the transfer. In the evening I locked the drawer and left.
Counting the log, it was twenty-three times.
I never wrote down why I opened it twenty-three times. When I try to write it, reasons come, but I doubt there are twenty-three of them. Mostly I would have opened it because it sat beside some other file I was looking for. Except that it does not sit beside any other file. Its registration number is nowhere near them.
I went to find the seven and open them. Finding them, there were eleven.
The four new ones were all private. One was a university club board, two were personal homepages, one was a page on a file-collection site. The collection site had trimmed the front and back and put up only twelve pages — and inside those twelve pages there were two lines that exist only in the file.
No copy had fewer lines. That was the same as last month.
What was different was the spread. Last month the copies differed in how many lines they held. This month they differed in how much they had gained, and the ones that had gained most were the ones whose views had risen most. A month ago I saw it by setting the two columns side by side. This time I knew before I set them side by side.
My hand stopped at the row for our registered copy.
Our registered copy is on the internal network only. It does not turn up in searches and cannot be opened from outside. Over the month, internal views of that file went from forty-eight to seventy-one, and twenty-three of those were mine. The log keeps employee numbers and times, so it comes out if you count. Everyone else's access falls in April and May. None in June.
Which is to say that for the month of June I was the only person who opened that file.
And that file gained a line.
I went and read it. The twelfth line. It sat in the same place and in the same manner as the eleven before it. A line that is not on the paper and is in the file only. Below the tide table in the lighthouse log, in what had been margin.
The line held a single tide. Date, time, height. The form matched the other rows of the table and only the figures differed. The date fell between the dates of the rows above and below it. It looked like a day that had been missing and was now filled in.
Read as a missing row filled in, there is nothing strange about it. The trouble is that this file was made by photographing the paper, and the paper does not have that row. I took the paper out again and looked at the place. Margin. It was margin on June 1 and it is margin today.
Last month I wrote down two readings. More lines drew more readers, or more readers grew more lines. Measure at two points in time, I wrote, and they separate.
They did not separate.
The second column does not erase the first reading. Of the copies other people read, it can still be said that they were read because they held more. But that cannot be said of our registered copy. Nobody read it and it gained a line, and nobody was not nobody. It was me.
To say that it grew because I read it, I would need grounds. To make the grounds, one copy would have to go a month without being opened. But to confirm that an unopened copy has not grown, I have to open it at the end of the month. Opening it, it can grow then. Not opening it, I do not know.
That there is no control to be had is what I confirmed in front of the second column.
A month ago I recognised the shape of this table, covered it, and had it open again within two hours. Today I did not cover it. I had already written down that covering changes nothing. Keeping to what one has written down and being unable to stop look the same from outside.
That passage in the survey log came back. The same place measured over and over, and the figure growing one way only. While scanning I read it as a record of error. I took it that the man measuring was unsteady and so the figures wavered. Wavering has no direction. The figures in that table had a direction.
Measure it and it deepens.
Put that sentence here and it comes out like this. Count it and it increases.
The surveyor took his rule down into the shaft; I sit at a desk and open a file. The work is the same. Today I understood what that man had been doing, and understanding it has not changed my hands.
In the afternoon I went round the eleven once more. Filling in the second column meant opening each of them anyway. Each time I opened one, the visitor figure went up by one. That one will go into the third column of this table next month.
The copy that had gained the most was the homepage where the text was typed out. The one where all hundred and eighteen pages were keyed in character by character. Seventeen to twenty-one, with visitors from fourteen thousand to twenty-three thousand.
The copies posted as photographs gained less. The copy that was written out gained more.
I noted the difference under the table separately. A photograph is something taken; a text typed out has passed through a hand once more. The one a hand passed through gains more.
Three of the four new lines were of the same grain as the earlier ones. Tides, depths, a sentence about something lit at night.
The fourth was of a different grain.
That line had a date and a page number. It was a sentence that set down figures without judgement. I know one person who writes that way.
My log stands at page sixty-eight today. The line gave page sixty-nine.
I read it three times. After the page number came a date, and after the date two figures. One of them was eleven. The other did not read. The characters were not smudged; the place was empty.
Our registered copy is on the internal network only. My log is in the drawer and the drawer is locked. One person has seen both.
Leaving work, I put the comparison table in the drawer. Laid over the access log, the twelve-line sheet, and the June table, it made four. Set side by side, the four are sheets that record four different things, and today all four looked like sheets that record one.
Tomorrow morning, when I open the log, it is page sixty-nine. I have not decided what to write.