Date: Tue, 21 Jun 2005 23:52:56 -0400 (EDT) From: "Keith F. Lynch" <kfl at KeithLynch.net> To: WSFA members <WSFAlist at WSFA.org> Subject: [WSFA] Re: Computer problems at work Sorry about the duplicate message last night. I carefully proofread the body of the text, but neglected to look closely at the header. I forgot one computer problem at work: Last week all new files in the remote directory on which I'm given my work were protected against my access. Within an hour, I had found the relevant Microsoft web page, printed it out, and given it to the people who were trying to give me the work. But they treated it like it was radioactive, or like I was deliberately making trouble for them, and asked me to do it at my end. Only today did the outside consultant get around to dropping by to fix the problem. Meanwhile, I had to log in as Ted. "Ted White" <twhite8 at cox.net> wrote: > Keith Lynch wrote: >> Yes, and those input plain text files look completely different: >> Every paragraph is on just one long line, there are no blank lines, >> there are no footers, there are no line numbers, and there are tabs >> before and after each speaker identification or Q or A. > Isn't it marvelous what Total Eclipse does with that? Sure. It would take me nearly a week to write a program that does the same. Much longer if I had to work under Windows, of course. > But there's a box to click off when saving Word documents as plain > text (.txt) which says something like "save line breaks." But that doesn't help me when saving *Total Eclipse* documents as plain text. If there's a way to get it to output files back into its input format, it's not mentioned in the manual. > (There's another box with something to do with character substitution, > which, since I know nothing about it, I leave unchecked.) I've discovered that it's a good thing to check. It turns wonky Microsoft-proprietary characters, such as those slanty quote marks, into proper ASCII. Total Eclipse is willing to take Microsoft- proprietary characters, but it's not happy with them -- its spell- checker catches them every time. And they will make any "ASCII" file made from the Total Eclipse file not a real ASCII file. >>> As I told you, you inherited Mary Catherine Gallagher's computer -- A Google search shows "Mary Catherine Gallagher" as a character on Saturday Night Live. > Mary could do fast, reasonably accurate work when she wished to, but > she spent most of her day surfing the internet, gossiping with other > Beta employees (we all heard all the dramas in her life -- and there > were many), or going out on errands. It sometimes took her a while > to realize that her machine had a new problem. You do know that "malingering" doesn't mean goofing off, it means pretending to be ill to get out of work? "Mike B." <omni at omniphile.com> wrote: >> Telnet is far older than Unix. > Yeah? Where did it come from then? It was developed for use on the ARPAnet, the predecessor to the Internet. The earliest reference I can find is RFC 96, dated February 12, 1971. That RFC doesn't mention Unix, which at the time was a non-networked proprietary in-house OS at Bell Labs. There were no Unix systems on the ARPAnet in 1971. > My impression was always that it was created as part of the Berkeley > work that led to TCP/IP and the whole "sockets" theory of network > programming. That would have been around 1970 and on what would > eventually become BSD Unix. TCP/IP dates to 1981; so does BSD Unix. BSD was designed to be fully compatible with TCP/IP, but TCP/IP was in no sense based on Unix. The whole idea of TCP/IP was to link together computers with different operating systems, different character sets, etc. That was also the whole idea of NCP a decade earlier. The main difference of TCP/IP is that it allowed the net to contain more than 254 computers, unlike NCP. >> Software registration is like countries -- the smaller and crappier >> a country is, the more draconian, slow, elaborate, and onerous the >> paperwork necessary to visit there. > Ever see what it takes to get into the USA? :-/ I rest my case. The idea of the USA as a citadel of freedom is dead. It's been replaced with the idea of the USA as a superpower and global policeman. They really ought to remove the Statue of Liberty, or at least the obsolete poem on it. > If reformatting in this way is your main use for the Eclipse > software, there are almost certainly cheaper alternatives that would > work better (i.e. without the bugs you describe). It's a full-featured word processor, very much like Microsoft Windows, and with a few of the same commands (e.g. control-C and control-V for cut and paste), but otherwise annoyingly different, especially when I have to switch back and forth every few days. For instance the arrow keys move by *words*, not characters. It knows, not just about characters and words, but also about the elements of structured speech found in depositions, i.e. there are questions by the lawyer, answers by the witness, colloquy, and various headers and parentheticals. They're all color-coded. And it's flagged as an error if there are two questions in a row without an answer between them, or two answers in a row without a question between them, etc. I could probably put together Emacs macros that would duplicate the functionality in a week or two. > I believe Word does have macros, but I've never used them. I > believe this because years ago there were a number of viruses that > spread due to the fact that Word was automatically executing any > macros embedded in a document when you opened the document... Yes, I heard about that. I'm just not sure whether those are macros in the same sense. Especially since Word documents seem to be all about *appearance*, while "Emacs documents" (i.e. plain text) are all about exactly what characters are present and in exactly what order. How does one formally deal with appearance? Two Word documents that look identical can be quite different internally, as people keep discovering to their dismay when text they thought they had deleted turns out to have still been present to anyone who looks at the internals. > I'm guessing that emacs has enough power in its macros to let you > turn that into your desired output format... Sure. It can do literally anything computable to a document. Also, it can edit absolutely any kind of file -- though if you don't know the file's internal format, you're sure to do more harm than good. I have edited .WAV files with it, though this is not recommended if you value your sanity. > Run emacs on Linux machines, or get the Windows version of > micro-emacs, Micro-emacs? Regular GNU Emacs runs on Windows. I installed it my first week there. > and you could eliminate those multi-thousand dollar licenses as > well as most of your bug-driven aggravations with Eclipse. I'm a *user* there, not a programmer. I will write these Emacs macros if my boss asks me to. But she won't. > It's also possible that your customers insist on this particular > software as part of the contracts... No. The customers are lawyers, who know nothing about computers. Customers never get Total Eclipse files, which would be of no value to anyone who doesn't have that program.