Home of the Boogie-Woogie Feng Shui

Redesigned ArtsMuse Website

The redesigned ArtsMuse web­site is live. This is the side project I am work­ing on with my wife and my friend Steve: a cus­tomiz­able art deliv­ery solu­tion for the home and office. What does that mean? It means we can put art on any TV!

Any­way, I redesigned our page. Below is a screen­shot of the new front page. The goal is to quickly get vis­i­tors to use­ful con­tent on one of the 3 land­ing sites for I want art, I make art, and I dis­play art.

Screenshot of revamped ArtsMuse main website page.

Screen­shot of revamped ArtsMuse main web­site page.

 

Jira Annoyances for Agile Tracking

We are using Jira more and more at work and I am find­ing it very frus­trat­ing. As an issue tracker it works well and links nicely with the other Atlass­ian prod­ucts we use, like Cru­cible. But now we are using it for agile iter­a­tion track­ing and it is just a stream of half-baked behavior.

Atlass­ian has put lots of effort towards cre­at­ing an ajax-heavy UI, but each action is as slow as a full page load. Then things that seem to be first-class arti­facts (Iter­a­tions) can’t be directly searched and linked-to with­out drop­ping into their SQL-like query syntax.

You have to get the actual iter­a­tion ID by click­ing the Iter­a­tion from within in a story and it changes from “iter­a­tion 1″ to some ID, in this case 2656. Then you can con­struct a search fil­ter based on that and fil­ter­ing out subtasks:

sprint in (2656) and issuetype in standardIssueTypes()

Then you can get a per­ma­nent link to it. They are pre­dictable after that, but for each iter­a­tion you have to locate this ID and con­struct a new query. There seems to be no Iter­a­tion His­tory type view or even a way to directly open them and look at them his­tor­i­cally. Iter­a­tion doesn’t seem to be a first-class con­cept, just a glommed-on bit of UI wrapping.

On Page Margins

Who decided 1.25″ was rea­son­able for mar­gins on paper? It’s stu­pid to waste 40% of the paper. I doubt peo­ple get con­fused when text comes to within say half an inch of the paper edge. “Gee, this para­graph would look bet­ter with a big white bor­der: like a mes­sage in a bot­tle drift­ing patiently in a vast sea wait­ing to be discovered.”

Here is the break­down of mar­gin reasonableness:

0 — Used by crazy peo­ple writ­ing screeds to hand to strangers on streetcorners

1/4″ — User clearly cares for envi­ron­ment but pushes the lim­its. Pos­si­bly goes a bit over­board about trees and mother nature.

1/2″ — The happy best size for com­mu­ni­cat­ing. LOOK — infor­ma­tion can be fit onto the page!

3/4″ — A good com­pro­mise for get­ting a lot of infor­ma­tion onto the paper with­out look­ing overcrowded.

1.25″ — Sta­tus quo. You write doc­u­ments but you don’t read them for a liv­ing or need to refer to them.

1.5″ — Gotta turn in a paper with spe­cific page count, eh? Hey, don’t for­get to use 16 pt font and make the mar­gins frac­tion­ally larger!

Greater than 1.5″ — You are liv­ing in a state of sin.

1.5″ mar­gins are barbaric.

Password-less, Key-based logins in Solaris

Found these notes in draft mode, writ­ten last sum­mer when I was doing heavy sysad­min work. These mostly call out impor­tant set­tings with­out a detailed set of steps. They may be use­ful for some­one though.


Users have to be set for no-password authentication.

bash-3.00# passwd -N jsmith
passwd: password information changed for jsmith

You can check by look­ing at /etc/shadow and ver­ify the pass­word col­umn is set to “NP”:

bash-3.00# grep jsmith /etc/shadow
jsmith:NP:15190::::::

 

Change sshd set­tings to dis­al­low Password-based logins

# Are logins to accounts with empty passwords allowed.
# If PermitEmptyPasswords is no, pass PAM_DISALLOW_NULL_AUTHTOK
# to pam_authenticate(3PAM).
PermitEmptyPasswords no

# To disable tunneled clear text passwords, change PasswordAuthentication to no.
PasswordAuthentication no

# Use PAM via keyboard interactive method for authentication.
# Depending on the setup of pam.conf(4) this may allow tunneled clear text
# passwords even when PasswordAuthentication is set to no. This is dependent
# on what the individual modules request and is out of the control of sshd
# or the protocol.
PAMAuthenticationViaKBDInt yes

# Are root logins permitted using sshd.
# Note that sshd uses pam_authenticate(3PAM) so the root (or any other) user
# maybe denied access by a PAM module regardless of this setting.
# Valid options are yes, without-password, no.
PermitRootLogin no

Short Digres­sion on spe­cial set­tings in /etc/shadow

Locked user

  • (not allowed to login)
  • All newly-created users are like this before you set a password
bash-3.00# grep jsmith /etc/shadow
jsmith:*LK*:::::::

No-Password Login: No password-based login

  • this is what you need for key-based logins
  • set with: passwd –N <user>
bash-3.00# grep jsmith /etc/shadow
jsmith:NP:15190::::::

[empty]: Dis­abled user

set with: passwd –d <user>

bash-3.00# grep jsmith /etc/shadow
jsmith::15190::::::

The Subway Experiment and Attention

I hate the con­clu­sion from this writeup of the sub­way exper­i­ment and the over­all tone. It’s an inter­est­ing exper­i­ment, but the con­clu­sion is irk­some non­sense. It assumes we should stop and eval­u­ate every­thing we see to deter­mine how great it is. Should every street musi­cian get 10 min­utes of time from every pedes­trian walk­ing by? No, of course not — who would be man­ning the bank teller or direct­ing air­plane traf­fic and the mil­lion other things that need to get done?

The rea­son tick­ets sell for $100 to see him in con­cert is our soci­ety has per­formed an effi­cient fil­ter­ing already — by pay­ing, a per­son is guar­an­teed to see some­one who is extra­or­di­nar­ily gifted and worth our lim­ited free time and money.

Also, the idea that this guy play­ing in the sub­way is the only pos­si­ble worth­while beauty nearby and every­one should stop and enjoy it is also silly. While walk­ing by, I can imag­ine the great din­ner with friends I had last night or the date night with my wife com­ing up, or lis­ten to the 1000 hours of self-selected great music from my phone. The idea some­body play­ing vio­lin should trump every­thing else vying for my atten­tion makes me angry: who is the author to judge what is best for me to think about while get­ting on the subway?