" /> General: March 2010 Archives

« February 2010 | Main | April 2010 »

March 06, 2010

Cell phones - Not up to snuff!

What is it with cell phones anyway. It's primary function, while I admit is not a function I do that frequently compared to other things, it can't do all that well. I mean calls but I don't mean hard to hear connections and dropped calls. Those are problems too however just the mechanism of placing the call...

First off - never make a fucking phone dialing app that is incapable of accepted a paste of a phone number!!! You'd think that'd be a no brainer but way too many phone dialers do not allow paste. So you find a number in a web page or an email or you put it in a file - whatever. You select the phone number and you go to the phone app and want to paste it in. There's no paste. Stupid phone!

Secondly you would think that a phone application would be, oh, I don't know.... RESPONSIVE PERHAPS. Many play back those button tones we all know and love. But most phones can't keep up with even the slowest of typists. How hard is this to sense the button was pushed and immediately (I mean right away, with out hesitation - IMMEDITATELY!!! - did I say immediately?) play the fucking tone and be ready for the next button push. Come on folks, this is not rocket science nor is this a lot of work that the phone needs to do in order to accomplish this minor fucking miracle - yet on both Windows Mobile based phones and now my Android based phone this constantly happens. Is it hardware related? Again, how much hardware is needed?!? My old land line never seemed to have an issue accomplishing this task - why can't my $$$ phone do it?!? Write the damn phone app in assembler and make it (and the tones for that matter) memory resident so you have no excuses WRT speed! Really is this too much to ask?!?

And another thing about phones - why is it that the most common things I do require push and hold in order to work. Not just phones, sometimes accessories. For example, my Bluetooth headset of choice (Sony Ericson HBH-980) has a recessed power button and you must hold for a few seconds to turn on. Now I understand the concept of requiring such a delay to distinguish between a real turn on/off event and an inadvertent one but trust me you would not be inadvertently hitting this power button.

And why is it that I must always wrestle with Bluetooth to get it to work right. My BT headset can handle both "Media audio" (from the media player) and "Phone audio" from say and incoming all. Why the distinction? I don't care. Just call it audio. You don't know how many time I try to connect my BT headset and only the phone audio gets connected. No media audio. So I try to connect again. No go. Try powering off the BT headset and back on. Nope. Try turning off BT on the phone and back on No go. Then the phone hangs. Bluetooth is the only thing that seems to have the power to hang my phone, requiring me to open the case and yank the battery in order to reboot!

And then when listening to Bluetooth often I'll get gaps of silence, especially when starting a new song or podcast, as if it needs to buffer stuff or whatever. It's real annoying. Why is Bluetooth so hard, flunky and trouble ridden?!?

Stupid is as stupid does

OK so I've moved up to Seattle and need to get my business and personal bank accounts all settled in. So I stopped into my local Wells Fargo branch to introduce myself. He said he could save me some money but unfortunately I need to get new accounts. With the new accounts come plain old boring cards. I had personalized my cards before. Now I have new plain cards I need to reactivate, only to have to re-personalize them again!

So I go to activate one of them and they ask for my 17 gillion digit card number then the last 4 numbers of my business checking account number. Well I don't know that so I need to go to the Wells Fargo bank site to get that. While I'm doing that it keeps saying "Sorry I didn't get that" and eventually I exceed my limit of retries. Hey I was just trying to get to your slow web site!!!.

Of course I'm dumped into the tech support queue where some ditzy blonde answers the phone. She tells me that she'll need to have my full business account number. I tell her I don't have that. I only have the last 4 because that's all your web page gives me. Normally I have to open up statement to get the full account number but I can't do that because this account is too new and doesn't have any statements. "Well you could just look at your checks..." to which I quickly interject "Listen sweetheart I haven't used checks in 18 years!". Then she tells me that she could help me but I'd need to give not one, nor two, nor even three but four forms of identification in order for her to help me! I told her that in no way would I be comfortable with that because everybody knows you don't achieve real security until after 20 forms of ID are presented and hung up!

OK stop and think here for a second... Why can the automated phone system validate me with just the mere last four digits of my account number, which is probably pretty damn easy to obtain, but the human needs the whole account number and/or 4 forms of ID?!?

Bottom line: Since I'm now at the web site I hang up on the dtiz and simply go back around and punch in my 4 digits and I'm validated. Stupid is as stupid does - and I ain't stupid - but I do see it all the time!

March 01, 2010

Priority Queues...

Many IT departments run long running and intensive processes on the various PCs throughout their organizations. For example, virus scanners (are evil but that's another rant), defragmentation and disk cleanup stuff. Is it really too much to ask for them to put such background tasks in the LOW PRIORITY QUEUE!!!. Do they know what the low priority queue even is or that it exists? I have yet to see anybody, other than myself, think ahead far enough to say "Gee this would interrupt interactive performance. I know! I'll put this in the low priority queue instead of being rude". I wonder what the thought process is for the IT guys or even if there is a thought process...

And then processes spawned from the original process go in the normal queue, so even if you put the first process in the low queue, the next process pops out in the normal queue. The correct way to do this is to have the original process, the parent process, in the low queue and have it insure any children created are also in the low queue!

Broken DNS

This seems to be the trend now a days in Corporate IT. Is it just Windows related? I don't know. It used to be that you could rely on DNS. Now you can't. Case in point - people hear talk about a machine known as "seacase". Others use the name usseacase. Whicih is correct? Well an educated guess would be they are one in the same and a DNS alias exists between them. IOW seacase is just a "pretty" name for usseacase.

Now I used to be that you could verify that with DNS, in particular you could use nslookup to find each of the names and see they have the same IP address. But more and more, Corporate IT DNS stuff is screwed up, perhaps in the name of security, such that they don't always resolve:

$ nslookup usseacase Non-authoritative answer: Server: <dns server> Address: 13.230.100.200

Name: usseacase.<domain>.com
Address: 13.230.100.74

So we can look that one up but we can't lookup the other:

$ nslookup seacase *** <dns server> can't find seacase: Non-existent domain Server: <dns server> Address: 13.230.100.200

However ping works! And verifies that this is the same machine. So why doesn't DNS know about this?

$ ping seacase PING seacase (13.230.100.74): 56 data bytes 64 bytes from 13.230.100.74: icmp_seq=0 ttl=127 time=230 ms 64 bytes from 13.230.100.74: icmp_seq=1 ttl=127 time=0 ms

----seacase PING Statistics----
2 packets transmitted, 2 packets received, 0.0% packet loss
round-trip (ms) min/avg/max/med = 0/115/230/115

Another pet peave of mine is people using either IP addresses instead of names (Do you refer to your friends by their cell phone numbers?) or using fully qualified machine names all the time when you don't need to...