I don't know about you, but using scp/pscp with Putty on Windows is always an eye-gouging experience. It's really what made me decide to setup a Linux computer. But I like Windows! So here's how I upload files our servers.

First, by the way, you should add it to your "path" environmental variable, so you can launch this shit from the command prompt. I did it by going to (My) Computer > Advanced > Environment Variables > System Vars. Select the "Path" and add this to the end:  "; C:\Program Files (x86)\PuTTY" or whatever your path to PuTTY is. Do it once and you can forget about it.

Maybe there is a way to use putty, but I don't know how. So you can open your command prompt. If you are using Windows 7, enter powershell, which is a smarter version of the command prompt. Actually, here's the whole deal, no more comments or further ado:

C:\Users\Scott>powershell
Windows PowerShell
Copyright (C) 2009 Microsoft Corporation. All rights reserved.

PS C:\Users\Scott>
PS C:\Users\Scott> pscp -P 2222 C:\Users\Scott\Desktop\modules.module scott@example.com:/tmp
The server's host key is not cached in the registry. You have no guarantee that the server is the computer you think it is.
The server's rsa2 key fingerprint is: ssh-rsa 2048 d0:75:70:12:db:23:e3:3a:b5:d9:64:bb:20:ce:ab:f4
If you trust this host, enter "y" to add the key to PuTTY's cache and carry on connecting.
If you want to carry on connecting just once, without adding the key to the cache, enter "n".
If you do not trust this host, press Return to abandon the connection.
Store key in cache? (y/n) y
scott@example.com's password:
modules.module     | 21 kB |  21.3 kB/s | ETA: 00:00:00 | 100%
PS C:\Users\Scott> exit

That's it. When I find an easier way I will post.

I have liked Good Magazine since before they were anchored to Starbuck's home page. I remember using it as a jumping off point for a few different websites almost 3 years ago. It's a great mix of content, design, and lots of charts. I sense there is a backlash against infographics, but I think anything that faithfully represents data in a compelling fashion is good.

git branch testing: new branch called testing
-v adds last commit
--merged shows branches merged into your currently checked out branch
--no-merged shows unmerged branches

Can't git branch -d (delete) unmerged branch without -D confirmation

git checkout testing: switches to that branch
git checkout master: switch back
git checkout -b iss53: create branch and switch to it

work needs to be committed before switching

to merge: go to master (or target branch) then merge branch into that branch
git checkout master
git merge hotfix
git branch -d hotfix: delete branch now that it is merged
(yep, merging doesn't merge everything, you still have the branch)

HEAD is like a pointer showing where you are on the commit/branch path

Merge conflicts:

GIT pauses; you can check status to see files needing resolution.

You go to the files and see:

Checked out code
========
Merging in code
>>>>>>>

Choose which you want or replace all. Add, commit and try again.
mergetool is graphical helper.

Remote Branches

If you clone, git calls this origin, points to master branch. Hence name: origin/master. This is a remote branch.
You also get your own master (local branch) to work off of.
Development can move ahead on the remote branch. To sync:
git fetch origin gets branches, but doesn't merge them. You do that:
git merge remote/branch (ex. origin/serverfix)
or, if you just want a working copy of the remote branch (they are read only)
git checkout -b alias remote/branch creates a new branch and moves into it. This new branch is a tracking branch, which allows you to push and pull to the remote branch. You can set this up for any branch.

git remote add alias_name path/to/repo.git adds a remote
git push remote remote_branch
git push origin :branch_to_delete deletes a remote branch

Rebasing

This is similar to merging but does it via a series of commits from the common ancestor of the branch, rather than a single code update. Allows for multiple branches to be committed in series. Replays commits.
This allows for skipping over a branch still in development.
git rebase --onto target_branch_to_replay_commits_to branch_to_skip_over (common ancestor) branch_to_replay_commits_from

Do not rebase commits that you have pushed to a public repository.

We have been working on our client onboarding process. You can probably look at each item and figure out what pain point I am trying to address. The biggest part of this process is managing client expectations: explaining why it is in their best interest that we don’t start working ticket the minute we shake hands. I have it roughly broken down into these steps:

1. Information Gather: Get all the access info we need at one time.
2. QA Test Criteria: How can our QA tell when the site is working? This is about collecting user cases we can put into selenium
3. Update Process: Are we doing updates? How often, who gets notified, etc.
4. Development Process: Are we the sole developers? If not, we either need to put in a release driven process with git process or just send a feature and let the client handle it.
5. Site Health Check: Run site analysis, code and server review
6. Task Assessment: Finally discuss in detail what the client wants to do

Internal Onboarding is next: Accounting/HR/Sysadmin: Add into systems, decide lead dev

There are no chapter one notes.

.gitignore can have rules:
# a comment - this is ignored
*.a       # no .a files
!lib.a    # but do track lib.a, even though you're ignoring .a files above
/TODO     # only ignore the root TODO file, not subdir/TODO
build/    # ignore all files in the build/ directory
doc/*.txt # ignore doc/notes.txt, but not doc/server/arch.txt

git diff shows changes not yet staged
git diff --cached or --staged shows staged but not committed
git commit -a -m 'comment'

git rm stages deleted files, removes from tracking & working directory (might need -f is you already modified and staged?)

git log shows commit history; add -p to show diff; -2 shows last 2
there are many formatting options for logging: git log --pretty=oneline

git reset HEAD -- filename unstages filename. Still modified (resets the index to what it was)
git checkout -- filename undoes modifications back to last commit; forever, no redo

git remote lists; -v shows url
git remote add alias url

git fetch pulls changes from remote; then you merge
git pull fetches a branch, pulls into current branch you are in
git clone adds remote repository as 'origin'. sets up a 'local master' to track 'remote master'

git push [remote-name] [branch-name]: when you want to share; only works if nobody has pushed in the meantime; otherwise pull first

remotes can be removed and renamed:
git remote rename pb paul
git remote rm paul

git tag shows all tags
git show shows annotated tag
git tag -l 'search term'
git tag -a 'tag' -m 'annotation'
git push origin [tagname] tags need to be explicitly pushed
git push --tags pushes all tags

Articles like this: "Big Data's Big Problem" make me mental. I have loved data analysis before Data Mining and Big Data became constant Linkedin feed fodder. I would really have to think about making such a big pivot, but working in SAS, hadoop, strata, etc. would be a blast.

http://www.quora.com/Career-Advice/How-do-I-become-a-data-scientist

Have some local files:
all_files/file_2.txt
file_1.txt

Locally create a repo, git init.
Create remote: ssh://scottco@174.1.1.1/~/public_html/git_test.git
Stage via "git add ." all those, commit. Ok, that works.

ssh onto remote server:
scottco@scottlawrencemassey.com [~]# cd public_html

This will be my bare repository, no website code, just git info goes here. I create the folder, cd into it, then git init --bare
scottco@scottlawrencemassey.com [~/public_html]# mkdir git_test.git
scottco@scottlawrencemassey.com [~/public_html]# cd git_test.git/
scottco@scottlawrencemassey.com [~/public_html/git_test.git]# git init --bare
Initialized empty Git repository in /home/scottlaw/public_html/git_test.git/

Seems to have worked...
scottco@scottlawrencemassey.com [~/public_html/git_test.git]# ls
./  ../  HEAD  branches/  config  description  hooks/  info/  objects/  refs/

So now on my local, I push to the bare repo on the remote. It worked!

Ok, let's get out of there and create the repo where the site will be.
scottco@scottlawrencemassey.com [~/public_html/git_test.git]# cd ../
scottco@scottlawrencemassey.com [~/public_html]# ls
./
cgi-bin/
robots.txt
../
.htaccess      
git_test.git/

scottco@scottlawrencemassey.com [~/public_html]# mkdir site_test
scottco@scottlawrencemassey.com [~/public_html]# cd site_test
scottco@scottlawrencemassey.com [~/public_html/site_test]# git init
Initialized empty Git repository in /home/scottlaw/public_html/site_test/.git/

Let's add a remote to the non-working tree
scottco@scottlawrencemassey.com [~/public_html/site_test]# cd ../
scottco@scottlawrencemassey.com [~/public_html/site_test]# git remote add git_test ../git_test.git/

What about adding a new file on my local: test_3.txt
Stage, Commit, Push. No error messages, think I am gtg, gtl, dtf.

The site needs that file too.
scottco@scottlawrencemassey.com [~/public_html/site_test]# git pull git_test master
From ../git_test
 * branch            master     -> FETCH_HEAD
Updating c1c2b17..f968927
Fast-forward
 0 files changed
 create mode 100644 file_3.txt

scottco@scottlawrencemassey.com [~/public_html/site_test]# cd all_files/
scottco@scottlawrencemassey.com [~/public_html/site_test/all_files]# ls
./  ../  file_2.txt  file_3.txt
Success!!

Pushing my luck:
Deleting file_3 from local, stage push
scottco@scottlawrencemassey.com [~/public_html/site_test/all_files]# git pull git_test master
remote: Counting objects: 5, done.
remote: Compressing objects: 100% (3/3), done.
remote: Total 3 (delta 1), reused 0 (delta 0)
Unpacking objects: 100% (3/3), done.
From ../git_test
 * branch            master     -> FETCH_HEAD
Removing all_files/file_3.txt
  UW PICO 5.05                      File: .git/MERGE_MSG

Merge branch 'master' of ../git_test

# Please enter a commit message to explain why this merge is necessary,
# especially if it merges an updated upstream into a topic branch.
#
# Lines starting with '#' will be ignored, and an empty message aborts
# the commit.

It merges, and asks for a commit message, but seemed to work:
scottco@scottlawrencemassey.com [~/public_html/site_test/all_files]# ls
./  ../  file_2.txt  

I think this is Tom Wolk on bass. It makes me happy to listen to him play, he played like a cross between Motown and Paul McCartney. Apparently, the song is about a dude with ties to Chicago, the son of the guy who brought KFC to Chicago. Yekh. Anyway, T-Bone Wolk is my favorite thing about Hall & Oates.

Just ordered real book; same price I paid 20 years ago.

Two great examples of space in music:

1. The clean guitar in “Another One Bites the Dust”
2. The bass line in the theme from “Mad Men”