Dubner can't be stopped! TapSIG is released!

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    OK. Got four new messages and just the four show in the "new" view.

    There is still the login issue with "Show in Forum" and Firefox as default browser.

    I will try it with Google Chrome as default browser next time around.

    I changed the application window position and size. The .ini file updated correctly and restart was correct.

    Again, my heartfelt thanks for this!
    Bacon is the answer. I forgot the question.

    Comment


    • #32
      Originally posted by Randy Sohn View Post

      Ohmigawsh, I've read this entire package in totality and must say that Bob, Ray, etc. could'a written all those entries in Greek and it would'a been just the the same to me - head shakin' mode here!

      best, randy
      Hah. Talk about old stuff -- Bob & Ray.

      I'll try it out after tax season.

      Comment


      • #33
        Ray, clicking the "Show in Forum" button causes me to launch a new process using the message's AVSIG URL as the command line. It's the same as if the URL were a file and you double-clicked it, or if you opened a command prompt window and entered "start http://dev.avsig.com/forum/hardware-software/2868-dubner-can-t-be-stopped-tapsig-is-released/page2"

        The CMD.EXE process notices the "http:" part of that and launches your default browser. TSCOM.EXE does use your username/password from the avsig.ini file to download messages directly -- no browser is involved, although I pretend to be "Mozilla/4.0", because otherwise at least some Web sites ignore me -- but the "Show In Forum" button launches your default browser, and from then on you're on your own.

        I only use Chrome, and somehow or other it seems to work with avsig.com so that I no longer have to keep logging in. For a while I seemed to have to log in every few minutes, but Mike O must've twisted the site's tail at some point.

        Comment


        • #34
          Bob, there is a "remember me" checkbox in the logon panel, but I assume it uses a long-term cookie. I scrub all cookies when I close the browser, so it wouldn't help much. Thanks for the explanation. It's what I figured. I also found out that if I log on to the first "show in forum", I get a second one OK, which is not what I thought was happening yesterday.
          Bacon is the answer. I forgot the question.

          Comment


          • #35
            **
            Originally posted by Ray Tackett View Post
            Bob, there is a "remember me" checkbox in the logon panel, but I assume it uses a long-term cookie. I scrub all cookies when I close the browser, so it wouldn't help much. Thanks for the explanation. It's what I figured.
            Then you'll be happy to know that in whatever folder you're running my thingie from, all the cookies are in the cookiejar.txt file. To my knowledge, nothing else is getting stashed anywhere.

            I can't think of any easy way of passing along the username/password pair to the browser. I'd have to go to the login page, and somehow cram the authentication data in. I ain't going to go that route.

            I am impressed by your functional paranoia, although I don't share it. I am troubled in principle by being tracked, but I am neither very troubled in that way, nor am I troubled at all in practice.

            There are just too many ways of being tracked. I don't want to scrub cookies, nor do I want to give up credit cards. And lately I've been mulling over the reality that most of us actually do have a defacto universal public identifier -- our cell phones. I am uniquely identified by my cell phone number. I use it for transactions. It broadcasts my location. It tracks where I've been. The browser history in the phone tracks at least some of my interests. There is a constant arms race now between me and organizations like Google, Facebook, the phone service and gawd knows who else who are highly motivated to snag that data from the phone and sell it to gawd knows who.

            I've learned to live with it.

            Comment


            • #36
              Thanks for the details. I'm not worried about storage of Avsig cookies, especially outside the web browsers' stores. Good luck tracking my flip phone on an account with web browsing and texting disabled. That's not security paranoia, just a belief that phones are for talking -- period.

              With browser fingerprinting, tracking is unavoidable, but it is possible to limit distribution of credit card information, not use PayPal, etc. I use one credit card online, the other in person. The "in person" card has been compromised four times, the online card once. My ATM card is never used as a debit card.

              A good ad blocker (Adblock plus or uBlock Origin) goes a long way toward avoiding web nasties and makes browsing far more pleasant.
              Bacon is the answer. I forgot the question.

              Comment


              • #37
                Originally posted by Ray Tackett View Post
                Good luck tracking my flip phone on an account with web browsing and texting disabled. That's not security paranoia, just a belief that phones are for talking -- period.
                You remind me of an article I read in the last month. Apparently some sociologists did a study demonstrating that by asking a few questions about cell phone usage, they can pin a respondent's age to within about five years.

                We share the house with our thirty year old daughter. If we need to communicate with her when she's not home, our best bet is a text message; almost invariably we get a response in a minute or so. But she hardly ever answers a phone call. And she says people ten years younger are even less likely to talk. Text, Twitter, Instagram, Snapchat... apparently all those, and more that I don't even know about, are used as active communications channels.

                Me? About one time in five, when I leave the house, I forget to bring the phone. It doesn't seem to be a problem. If it's a real emergency, you need to call 9-1-1, not me.

                About seven years ago, we had a spate of credit card compromises, mostly showing up as tiny charges from countries in Southeast Asia. Happily, nothing big ever happened after those, "Does this number work?" test charges. Judy got annoyed, and asked American Express to give her five cards, all on the same account but with different numbers. AmEx was reluctant to do that, but Judy persisted and they acquiesced.

                By divvying up what cards she used for what purposes, we isolated it to a particular Japanese restaurant near 116th Street and Broadway. We mulled it over, and then shrugged it off and decided on a simple policy. Cash only for takeout, instead of reeling off a credit card number over the phone to people with every motivation for selling those numbers. That was the end of the problem.

                Comment


                • #38
                  Originally posted by Robert Dubner View Post

                  we isolated it to a particular Japanese restaurant near 116th Street and Broadway.
                  Hope it wasn't the old Ollies before it burned down....

                  Comment


                  • #39
                    Okay, all you TapSIG users. (As far as I know, that's Ray, but I have to assume there are multitudes.) I have released Version 1.1 at http://www.dubner.com/tapsig. As far as I am concerned, it's complete. At the present moment I have no additional development objectives for it. The one current glitch that I am aware of is that sometimes, after a quoted segment, the following text starts on the same line as the quote, rather than starting on the line after the quote.

                    To fix this means implementing a "conditional newline", which would probably take me about four minutes, but then I'd have to spend another fifteen minutes updating and double-checking the .ZIP release packages...and I don't feel like it. And it's not a big problem, because it really is doing exactly what I want it to be doing.

                    Again, if you play with it, let me know. A number of people have told me they aren't using it, which is kind of useful, although if all seven billion people who aren't using it chime in Mike O is going to start to run into storage problems.

                    But now I am going to work on something else. I am going to build a thermometer.

                    Comment


                    • #40
                      I downloaded and extracted 1.1. It seemed to reindex the existing message base when it first started. Other than that, I see a new icon, but cannot find anything which says what's new in this version.

                      I will have to wait for some new messages to come in to test the get new function. Thanks again for doing this. While the new forum is easier to navigate via web browsing, Tapsig showed me how much new stuff I was missing. The clear distinction of new (to me) messages is valuable.
                      Bacon is the answer. I forgot the question.

                      Comment


                      • #41
                        Originally posted by Ray Tackett View Post
                        The clear distinction of new (to me) messages is valuable.
                        How valuable?

                        (Just getting you a new message.)

                        Comment


                        • #42
                          Originally posted by Ray Tackett View Post
                          I downloaded and extracted 1.1. It seemed to reindex the existing message base when it first started. Other than that, I see a new icon, but cannot find anything which says what's new in this version.
                          Actually, TapSIG scans the entire message file every time you start it and every time you download new messages. The change that you noticed is that 1.1 sorts the sections into the same order that Mike O. built into the Website. That is, People/Places/Gigs comes first, followed by Hangar Flying, and so on.

                          I made a bunch of changes that are, if I was successful, going to be hard to see, because they were done to make the text box look as much like the forum as I easily could. The big exception is embedded graphics -- I don't even try.

                          But the alpha version didn't even put up the hyperlinks. Now it does. The exception to that are smilies. I suppress the hyperlink for smilies, and just put up text: shows up as <Stick Out Tongue>, or something like that.

                          Other things: Ctrl+F starts a forward-only find from the beginning of the message file. F3 and 'N' continue the search. It searches the hypertext, so you might find messages showing up as 'found' without any visible reason.

                          The 'H' key now toggles a 'hold' feature. Messages flagged as 'held' show up in both the Unread and New views.

                          At some point I'll update the one-page user manual.

                          And, yes, I replaced Microsoft's default "MFC" icon with the venerable Dubner Computer System's triangular Möbius strip. Gads, that graphic goes back over forty-five years.

                          Keep me posted about things that don't work the way you think they should.
                          Last edited by Robert Dubner; 04-09-2018, 23:55.

                          Comment


                          • #43
                            Hmmm. I apparently have some kind of problem that pops up when I try to download new messages and there aren't any.

                            Edge cases. Corner cases. Drive you nuts!

                            Okay then. Version 1.2 is up. My advice: Stick with UNREAD rather than NEW. If you click on the upper-left-corner icon, you have the option of marking everything as READ, which means, going forward, you can click Fetch New to your heart's content without missing anything. The NEW view is just that; it just shows you stuff since the last time you clicked Fetch New. If there isn't anything new, those bits get cleared and you can't get them back. UNREAD is the hot ticket, as far as I am concerned.
                            Last edited by Robert Dubner; 04-10-2018, 10:11.

                            Comment


                            • #44
                              Edge cases. Corner cases. Drive you nuts!
                              Indeed. I hit that case last night. All messages showed up as new. The log showed there were no new messages. Apparently the code doesn't expect an empty view. That's fine, but it may be worth a documentation note if you're updating the doc anyway.
                              Bacon is the answer. I forgot the question.

                              Comment


                              • #45
                                Originally posted by Ray Tackett View Post

                                Indeed. I hit that case last night. All messages showed up as new. The log showed there were no new messages. Apparently the code doesn't expect an empty view. That's fine, but it may be worth a documentation note if you're updating the doc anyway.
                                Ray, I had just goofed. What I had in mind was that if there were absolutely no new messages, I'd just leave the NEW flag alone and leave the prior set of new messages still flagged as new. But that meant I needed to distinguish between messages that were new and messages that really are brand-new after a forum read. That meant I not only have a NEW flag, I also have a BRANDNEW flag that isn't visible to you, the user.

                                But my mistake was something this: When I read the forum messages with a CTRL+N, my code first carefully clears out all the BRANDNEW flags before reading the forum and then scanning the message file. But when I fired off the program initially, I was scanning the message file without first clearing the BRANDNEW flags (because I wasn't reading the forum). And because my message object constructor initially defaults the BRANDNEW flag to ON, the result was an execution path that ended up with all the NEW flags on.

                                A little more thought made me realize my initial design goal was ill-advised. If you ask to see NEW messages, and there aren't any, then you should end up seeing no NEW messages, not see a repeat of the last download where there were new messages. So, the code is now simpler; I am not doing different things depending on whether or not there were any new messages.

                                Besides, I don't know how necessary the *New* view is, anyway. Stick to *Unread* in normal use.

                                Comment

                                Working...
                                X