Tuesday, August 27, 2013

CPAN: Terminal does not support GetHistory (No ReadLine support)


Surprised to see these messages come up again, I thought this had gone away in the 90's.

Disk is cheap now; why not just include all these packages?

Or at least:  grab my personal [developer] profile, see all the packages I have installed,
offer to display them for me.
Show me which I use frequently, a web page with statistics and links to latest news...

In case I come across this again, I will note:
CPAN:
Terminal does not support GetHistory
Terminal does not support AddHistory


if you are using a 'straight' CPAN system:

install Term::ReadLine::Gnu


linux, as a package:
sudo apt-get install libterm-readline-gnu-perl 

Sunday, August 18, 2013

Annoyances for 18 August, 2013

Actually, these have been building for a long time; I've just decided to collect them now, and ideally, find some answers.

Bitcasa:

  • Opaqueness of Bitcasa:  it says "513kb to go" or "788 bytes to go", and seems to just switch back and forth between those numbers.  Network bandwidth at 250KB/sec for several minutes, never getting to zero.  3 hours to finally finish from where it said 13M (and 33 hours from ~1570Mb...
  • Efficiency of Bitcasa:  800M of RAM when I'm just looking at files?  1G+ sometimes?  You're using a huge chunk of disk for cache, what do you need all the RAM for ?
  • Interface:  I alt-tab over, Cmd-, to bring up preferences (only to see how munch I am using) and then what key do I use to get rid of the window?  ESC?  Cmd-W ?
    No, I HAVE TO CLICK.  I DON"T WANT TO USE THE MOUSE>
  • bitcasa, why 15 minutes to list 600000 files (using find + stat)?
    Doulbe why:    sustained CPU at 80% during this time, but not much network.  Should be less that 50M total transmitted (need to estimate).  I saw stuff disappear before; until I trust you more, I'm going to keep checking.  [Maybe staggered checking is better.]


Last.fm -- 25M download?  It seemed to never install successfully either.  Finally, with iTunes 11 it does?

iTunes -- you say you're started, but no window?  Why don't windows show when I alt-tab to the icon?  Where are you?  I have to click ?!?  To get the 'agreement' window for new iTumes?  what's so special about this window, why can't it act like a normal window?

Maybe it's not just iTunes, maybe other applications too since Mountain Lion.  Or maybe just Apple's apps ?

Mountain Lion (10.8.4) - I finally recently upgrade from Snow Leapord (10.6.8) about 3 weeks ago--had to because my LEAPmotion controller required it.
Pros:

  • Upgrade process *very* smooth -- totally online, cheap ($20), no effort (1 click, walk away, come back, no surprises)
  • Hibernate seems to work now; at least it has a couple times.  I frequently fall asleep on the couch after opening my laptop when it's charged, and don't bother plugging in (or it comes unplugged)...  then I wake up to it being off.   Maybe I should take care of the "Service battery" status?  -- but pretty impressed it still gives 2 hours after 4 years.
  • Re-opens (all?) applications that were running from before reboot.
    Slightly mixed about this; Terminal opens, but not all my screen windows.  
  • TimeMachine also uses free disk when you're not connected to your TM backup drive?
    or is this part of MobileMe, I mean iCloud ? no, I didn't sign up.
  • Also, java:  automatically downloads 1.6 when I started Eclipse;
    and system updates automatically install 1.7 (aka Java 7).
  • XCode (developer tools, compiler etc) and easy one-click install from the app store.
    Mostly I needed it for 'macports' so I could update it (because my old system was so out of date..).
    Possibly useful in the future for iOS, though I think 1) Android is bigger and better, and 2) HTML5 is a better choice before native iOS.  Still, a think iOS wrapper app might be able to provide some control/integration features.
  • Notification center -- seems to be a good centralized idea.
    Configurable.  Note to self:  I should maybe not have my calendar reminders on screen ?
Cons  (or more things that I haven't gotten used to (yet)):
  • 'Natural scrolling'
  • 'desktops' v spaces.  wish I could rename.
    but maybe it best that they remain temporary, to reduce multi-tasking ?
  • CalendarAgent takes a lot of cpu, never syncs...
  • MagicPrefs starts with two icons (doesn't seem to be two copies running)
  • Bitcasa *doesn't* autostart.
  • VMware .. gone.  I didn't upgrade before even when VMware fusion nagged me;
    after the upgrade, it's "damaged goods".
    Maybe I'll just stick with VirtualBox -- it's multi-platform.
  • no Rosetta?  I might like for a thing or two .. ?
  • app store.  I'm an old-school sysadmin control-freak who wants total control;
    maybe I should just relent, realize you can't pop the hood on some machines
Still note this is WAY ahead of WINDOWS.  To people who complain Macs are more expensive, realize that you will not have to spend 1-2 hours per week with anti-virus scans; your disk speed is not reduced because every file also has to be read by anti-virus software before your program starts or your application can read it.  How much is your time worth to you?  Consider the reduced frustration and stress.  (Windows is a sinking ship ..).  

windows abomination legacy sinking ship; please, launch your torpedo.  competition is good, but windows is just not up to the task.  everyone, just keep it in some virtual corner (if you have to have it) as a warning to future occupants.  suckit ballmer.

LEAPmotion - eats a ton of cpu.  20% when no movement, 65% cpu when I'm moving my hand around it, even when nothing is reading...  (so I have an old computer..  2.5GHz dual core should not require that much).

VMware -- something better could be done than being gone after ML upgrade.  Is it ML's fault?

CiscoVPN -- fills up my log files (system.log, others?) with messages like
acwebsecagent[151]: License : One or more of the License/Public Key can't be NULL
and plenty others.
[ maybe uninstall, re-install w/ out certain components ? ]



General notes on software:
although better than ever, the current state is still lagging behind my vision by what seems like five, ten years, and growing.  So while I get older, more jaded, more cynical and curmudgeonly, it's more frustrating.  

On the other hand, I understand how hard it is to write software..  and how short our vision really is, a limited understanding and capacity for anything beyond a few chunks at a time.


On the third hand, there is more development than ever before, exponentially so; I get frustrated with myself trying to keep up.

There are better software management tools.  There are just so many.  And each are different, for
rpm's, pkg, npm, ruby gems, python libs, github, maven, CPAN, JPAN, RCRAN (?);
sourceforge, google code,



... will organzine and update more later.
maybe find some answers.

Wednesday, August 7, 2013

Is making the Web look like paper the best we can do ?

Why hypertext was better before WWW.

[not sure if pro & con, or a quadrants (2x2 ) are better.   I think there's only 2 choices..
hmm..
what if a tool allows you to pick importance of issues, and weights
and compare your graph to others?

or (better?)  query before, and then after you read to guage any change in attitude.  then show you the graph (and others graphs)


pro pre-

  • magical potential
  • very different from paper
  • no interface to get in the way;  no browser quirks, no ads!
  • semantic, typed links
  • texts were all-encompassing, contained everything you needed but showed you only what you wanted
  • Xanadu and Ted Nelson was less jaded
con pre-
  • mostly fictional
  • each hypertext a silo;  completely different models, tools, architecture


IBIS as argument store.
Nodes in the graph can represent attitudes

.. a work in progress ..

Sunday, August 4, 2013

Macbook external monitor loses sync, goes into loop flashing blue

Symptom:

External monitor (connected via mini-port to hdmi adapter) loses sync or doesn't display, Macbook screen flashes blue like it wants to show the external display but can't.  Keeps doing this over and over again.

The first time I plug it in it nearly always gets the connection.  The time that it fails seems to be when already plugged in, and the screen wakes up from screen saver.

Solution:

Unplug and plug mini-port display adapter back in.  This works for me at least 95% of the time the first time, then probably 100% of the time the second time.  Unless the adapter has failed, in which case it won't.

When the adapter is failing, there is no way to determine what is going on.  Results will be intermittant and random, and (can't be) explained by any event.  Might be the phase of the moon, could be sacrificing a goat or chicken, and then the next day it's the chicken.  Or the goat.  It will be just like religion.

I wish there was a way to reset it; I have tried the following:

  • like put the monitors back to sleep, then re-awaken.  This doesn't help.  
  • I've also tried putting the computer to sleep, then waking back up.
  • Bring up the monitors controller, while the screen is flashing from blue to normal,
    and trying "Detect displays" but this hasn't worked
  • Turning the external monitor off and back on.
    When it's off, it behaves exactly if there is no second monitor connected.
    When on, it goes into flashing blue mode again.
    -->  this implies it is in the state of the Mac, not the state of the monitor.

    To TRY:
     - put mac to sleep, disconnect adapter and plug back in, then wake up mac
There is also "resetting the parameter RAM" and SMC
(doesn't help if adapter is broken).

What worked for you ?

2013.08.28 Update:
After upgrading to Mountain Lion 10.8.4 (previously was Snow Leopard, 10.6.8) it seems it might be a little better.  Some differences:

  • After wake from sleeping display (not awake from standby, just after the screen has blanked) the laptop may have a blue screen but the external monitor remains blank and powered off.  After entering password, the laptop screen came back (no longer blue);
    once the external monitor came back, another time it didn't (I had to unplug/plug it back it to get the display back).
  • A couple times my system seemed to hang completely after the display woke back up, getting hot and with the fan running, to the extent I had to forcibly power it off (it may have already been hung to some extent due to other processes, but the display partially "woke up").

    Once I was able to log in remotely to see was might be going on, first thing I saw was
    21:53  up 1 day,  4:39, 10 users, load averages: 66.81 27.60 11.78
    (note that is an EXTREMELY high load average) and saw these processes:


   88    92     1   0 Mon05PM ??        30:22.22 /System/Library/Frameworks/ApplicationServices.framework/Frameworks/CoreGraphics.framework/Resources/WindowServer -daemon
  501 16972    92   0  9:53PM ??         0:00.25 /System/Library/CoreServices/DMProxy -discovery
  501 16973    92   0  9:53PM ??         0:00.24 /System/Library/CoreServices/DMProxy -login
    0 16974    92   0  9:53PM ??         0:00.27 /System/Library/CoreServices/DMProxy -login
    0 16976    92   0  9:53PM ??         0:00.22 /System/Library/CoreServices/DMProxy -discovery
  501 16977    92   0  9:53PM ??         0:00.19 /System/Library/CoreServices/DMProxy -login
  501 16978    92   0  9:53PM ??         0:00.19 /System/Library/CoreServices/DMProxy -login
  501 16979    92   0  9:53PM ??         0:00.19 /System/Library/CoreServices/DMProxy -login
  501 16980    92   0  9:54PM ??         0:00.07 /System/Library/CoreServices/DMProxy -discovery
  501 16981    92   0  9:54PM ??         0:00.07 /System/Library/CoreServices/DMProxy -discovery
    0 16982    92   0  9:54PM ??         0:00.08 /System/Library/CoreServices/DMProxy -discovery
  501 16999    92   0  9:54PM ??         0:00.04 /System/Library/CoreServices/DMProxy -discovery
  501 17000    92   0  9:54PM ??         0:00.03 /System/Library/CoreServices/DMProxy -discovery
  501 17001    92   0  9:54PM ??         0:00.03 /System/Library/CoreServices/DMProxy -discovery


I unplugged the miniport adapter, and everything started to go back to normal.  Lots of activity in the log files (/var/log/system.log) but I haven't analyzed in detail yet, but messages like this repeat over and over:

Aug 20 21:49:38 macbookpro WindowServer[92]: Received display connect changed for display 0x4272900
Aug 20 21:49:38 macbookpro WindowServer[92]: Created shield window 0x29c9 for display 0x04272900
Aug 20 21:49:38 macbookpro WindowServer[92]: handle_will_sleep_auth_and_shield_windows: no action for lock state 1
Aug 20 21:49:38 macbookpro WindowServer[92]: Received display connect changed for display 0x3f003d
Aug 20 21:49:39 macbookpro WindowServer[92]: Display 0x04272900: GL mask 0x1; bounds (0, 0)[1440 x 900], 36 modes available
Aug 20 21:49:39 macbookpro WindowServer[92]: GLCompositor: GL renderer id 0x0102260e, GL mask 0x00000003, accelerator 0x00003c7f, unit 0, caps QEX|QGL|MIPMAP, vram 256 MB
Aug 20 21:49:39 macbookpro WindowServer[92]: Display 0x003f003d: GL mask 0x2; bounds (2464, 0)[1 x 1], 1 modes available
Aug 20 21:49:39 macbookpro WindowServer[92]: Created shield window 0x29ca for display 0x04272900
Aug 20 21:49:39 macbookpro WindowServer[92]: Created shield window 0x29cb for display 0x003f003d
Aug 20 21:49:40 macbookpro WindowServer[92]: **DMPROXY** (2) Found `/System/Library/CoreServices/DMProxy'.


Aug 20 21:49:44 macbookpro WindowServer[92]: post_notification : Time out waiting for reply from "Flux" for notification type 109 (CID 0x16603, PID 992)
Aug 20 21:49:44 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0: Startup Mode 1440 x 900, CGSThirtytwoBitColor, Resolution 1, IOMode 0x80000000, IODepth 0x2, IOReturn 0x0
Aug 20 21:49:44 macbookpro WindowServer[92]: Display added
Aug 20 21:49:44 macbookpro WindowServer[92]: Display removed
Aug 20 21:49:44 macbookpro WindowServer[92]: Display 0x04272900: GL mask 0x1; bounds (0, 0)[1440 x 900], 36 modes available
Aug 20 21:49:44 macbookpro WindowServer[92]: GLCompositor: GL renderer id 0x0102260e, GL mask 0x00000003, accelerator 0x00003c7f, unit 0, caps QEX|QGL|MIPMAP, vram 256 MB
Aug 20 21:49:44 macbookpro WindowServer[92]: Display 0x003f003d: GL mask 0x2; bounds (2464, 0)[1 x 1], 1 modes available
Aug 20 21:49:44 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:49:45 --- last message repeated 1 time ---


Aug 20 21:49:46 macbookpro WindowServer[92]: **DMPROXY** (2) Found `/System/Library/CoreServices/DMProxy'.
Aug 20 21:49:46 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:49:48 --- last message repeated 1 time ---


Aug 20 21:50:51 macbookpro WindowServer[92]: Received display connect changed for display 0x4272900
Aug 20 21:50:51 macbookpro WindowServer[92]: Created shield window 0x2a18 for display 0x04272900
Aug 20 21:50:51 macbookpro WindowServer[92]: handle_will_sleep_auth_and_shield_windows: no action for lock state 1
Aug 20 21:50:51 macbookpro WindowServer[92]: Received display connect changed for display 0x3f003d
Aug 20 21:50:51 macbookpro WindowServer[92]: Created shield window 0x2a19 for display 0x19c6b0c1
Aug 20 21:50:51 macbookpro WindowServer[92]: handle_will_sleep_auth_and_shield_windows: no action for lock state 1
Aug 20 21:50:52 macbookpro WindowServer[92]: Display 0x19c6b0c1: GL mask 0x2; bounds (0, 0)[1920 x 1080], 48 modes available
Aug 20 21:50:52 macbookpro WindowServer[92]: Display 0x04272900: GL mask 0x1; bounds (-1440, 0)[1440 x 900], 36 modes available
Aug 20 21:50:52 macbookpro WindowServer[92]: GLCompositor: GL renderer id 0x0102260e, GL mask 0x00000003, accelerator 0x00003c7f, unit 0, caps QEX|QGL|MIPMAP, vram 256 MB
Aug 20 21:50:52 macbookpro WindowServer[92]: Created shield window 0x2a1a for display 0x19c6b0c1
Aug 20 21:50:52 macbookpro WindowServer[92]: Created shield window 0x2a1b for display 0x04272900
Aug 20 21:50:52 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1: Startup Mode 1920 x 1080, CGSThirtytwoBitColor, Resolution 1, IOMode 0x80001000, IODepth 0x2, IOReturn 0x0
Aug 20 21:50:52 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0: Startup Mode 1440 x 900, CGSThirtytwoBitColor, Resolution 1, IOMode 0x80001000, IODepth 0x2, IOReturn 0x0
Aug 20 21:50:52 macbookpro WindowServer[92]: Display added
Aug 20 21:50:52 macbookpro WindowServer[92]: Display removed
Aug 20 21:50:52 macbookpro WindowServer[92]: Display 0x19c6b0c1: GL mask 0x2; bounds (0, 0)[1920 x 1080], 48 modes available
Aug 20 21:50:52 macbookpro WindowServer[92]: GLCompositor: GL renderer id 0x0102260e, GL mask 0x00000003, accelerator 0x00003c7f, unit 0, caps QEX|QGL|MIPMAP, vram 256 MB
Aug 20 21:50:52 macbookpro WindowServer[92]: Display 0x04272900: GL mask 0x1; bounds (-1440, 369)[1440 x 900], 36 modes available
Aug 20 21:50:52 macbookpro WindowServer[92]: GLCompositor: GL renderer id 0x0102260e, GL mask 0x00000003, accelerator 0x00003c7f, unit 0, caps QEX|QGL|MIPMAP, vram 256 MB
Aug 20 21:50:53 macbookpro WindowServer[92]: **DMPROXY** (2) Found `/System/Library/CoreServices/DMProxy'.
Aug 20 21:50:55 --- last message repeated 3 times ---
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1; ColorProfile { 4, "HF237"}; TransferFormula (1.000000, 1.000000, 1.000000)
Aug 20 21:50:55 --- last message repeated 1 time ---
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:50:55 --- last message repeated 1 time ---
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1; ColorProfile { 4, "HF237"}; TransferFormula (1.000000, 1.000000, 1.000000)
Aug 20 21:50:55 --- last message repeated 1 time ---
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1; ColorProfile { 4, "HF237"}; TransferFormula (1.000000, 1.000000, 1.000000)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1; ColorProfile { 4, "HF237"}; TransferFormula (1.000000, 1.000000, 1.000000)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1; ColorProfile { 4, "HF237"}; TransferFormula (1.000000, 1.000000, 1.000000)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:50:55 --- last message repeated 1 time ---
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x19c6b0c1: MappedDisplay Unit 1; ColorProfile { 4, "HF237"}; TransferFormula (1.000000, 1.000000, 1.000000)
Aug 20 21:50:55 macbookpro WindowServer[92]: Display 0x04272900: MappedDisplay Unit 0; ColorProfile { 3, "Color LCD"}; TransferTable (256, 3)
Aug 20 21:51:17 macbookpro WindowServer[92]: Received display connect changed for display 0x4272900
Aug 20 21:51:17 macbookpro WindowServer[92]: Created shield window 0x2a47 for display 0x04272900
Aug 20 21:51:17 macbookpro WindowServer[92]: handle_will_sleep_auth_and_shield_windows: no action for lock state 1
Aug 20 21:51:17 macbookpro WindowServer[92]: Received display connect changed for display 0x19c6b0c1
Aug 20 21:51:17 macbookpro Flux[992]: CGSGetSizeOfDisplayTransfer: Invalid display 0x19c6b0c1
Aug 20 21:51:17 macbookpro Google Chrome[1011]: CGSGetCurrentDisplayMode: Invalid display 0x19c6b0c1
Aug 20 21:51:17 --- last message repeated 1 time ---
Aug 20 21:51:17 macbookpro Terminal[313]: CGSGetCurrentDisplayMode: Invalid display 0x19c6b0c1
Aug 20 21:51:17 --- last message repeated 1 time ---
Aug 20 21:51:17 macbookpro Flux[992]: CGSGetSizeOfDisplayTransfer: Invalid display 0x19c6b0c1

... Lots of 'display added', 'display removed' messages;  oddly, the external monitor (HF237) was not identified until about 3 minutes later (but still came as invalid).

Anyway, unplugging and plugging the display back in still seems to be the best solution


2014 Update:  after updating to OS X 10.8 (Mountain Lion) this has rarely happened.  Occasionally the external doesn't properly reconnect after display wakeup or resume from standby, but not the endless looping with screen flashing blue.  Unplug/plug back in always seems to fix it, it's just required a lot less frequently.