CodeDog.Net

UPDATE : 10/10/2012 8:00am : Apparently I have no idea what I’m doing :-/  I built the tutorial as I was building the scripts. These did not actually work at the end of the day or the next morning.  Working on determining why.

UPDATE : 10/10/2012 8:45am : So I figured out these do indeed work, however, my Automator Launcher process was wedged, and these three scripts were just sitting there, not doing anything.  I found this by running Activity Monitor and looking at the processes.  I killed off my scripts and the launcher process, and now they work.

I’m lazy.  If I can write a program or script to do something to save me time, I’ll do it. Especially when I have to do something over and over and over. Take for example connecting to servers every morning when you log in, or disconnecting them when you leave at night.  I have a MacBookPro and I repeatedly connect and disconnect to various places on campus.  I also have a fire-wire drive that is my Time Machine backup.  If you’ve used a Mac, you know you get ‘yelled at’ if you simply unplug it without properly dismounting it.

When you have 4-5 servers connected and external devices connected, that’s a lot of wasted time even just ejecting them when you want to leave. So I started a quest to save me time (even if its just 2 minutes a day, it adds up, right?).

Enter Automator.  This is a special utility on Mac that lets you build a set of steps and run it.  These can be as simple or as complex as you want.   Here is what I wanted to accomplish –

• connect to my 5 servers at 7:30 (when I’m typically at my desk)
• disconnect from those servers at 4:50 (when I’m ready to leave)
• disconnect from my fire-wire drive at 4:50 (since this auto-mounts when I plug it in in the morning)

Here’s what I did –

Open Automator (from the Launchpad)

New Application

On the left is a set of action.  Either choose the Files & Folders listing from the Library column, or type in Get Specified Servers.

Drag that from the left onto the right pane like the image below –

Then we want to tell Automator to do the actual connecting to these servers.  Be advised – if you have to log into these servers, and those credentials are not stored in your key ring, it will prompt you to log in to each server.

Now save it.

Create a new iCal Alarm – File | New | iCal Alarm

Find the Launch Application action and drag it over.

Choose ‘other’ from the drop down list and go find where you saved the application above.

Choose the application you built before and click Choose. Along with any other actions  you want. I build a separate application to disconnect my Time Machine drive, and added it below.  Save this and it will prompt for an alarm name –

Once you enter that, iCal will launch and you can then edit the alarm details, such as when time and how often you want this to run – for me we’ll set this to every day @ 7:30am –

Now my computer will connect every morning and save me the headache of doing it myself (I said I was lazy, remember)

So now – at the end of the day – I have to pack up and clean out my coffee pot – so I want the computer to disconnect from all of these things.  Earlier I created a separate application to disconnect from my Time Machine drive, so I could execute that separately if I had to.  I’ll set up a new iCal Alarm to disconnect the servers and my drive at 4:50PM each day  just like we did before.

then save it –

Then set the time / date in iCal when it opens.

Now you’re done, and I’m happily lazy.

I’ve recently completed a web app for an event registration on campus, and it is now being tested by the client. I sent an email to the client letting them know its ready for them to test it and to send me any changes they would like to implement (I get a salary so doesn’t matter if is 1 thing or 100 things).

They actually did a nice job going through and making comments, however, their delivery put me off a bit.

Why will people say – “The app [insert whatever here] works great [ sometimes they even say perfectly] and we’re ready to launch it.  Can you change this text, change this email subject line, make the user experience path go here instead, alter these prices and add a link to our Facebook page?”

Ok.  All really good feedback (because the customer is always right, except for when they’re incredibly stupid), but don’t tease me with the first line into a false hope that I’m “done” working on the project.

I’ve done this enough to know – once it launches there will be tweaks, that’s expected.  I understand and HOPE the client tests (beyond my testing) it and makes comments and had changes.  But please don’t tell me its ready to launch – and give me a pick list like a housing inspector.  “Yea, this house is ready to sell, you just need to replace the roof, fix the railings on the deck, patch the sidewalk, repair the furnace”

I’ve also done this long enough to expect this type of to do list, today it just got to me.  Maybe they’re trying to make me feel good before dumping work on me.  Maybe they’re uncomfortable asking me to change things, I’m not sure. I’m happy to change things, I even asked them to tell me what I needed to change.  But that “we’re ready to launch” followed by the task list thing stuck me.

Doing that completely negates the first sentences of the email where you praise and assure me you’re happy with the solution.  I’m all about improving something I’ve built ESPECIALLY when it improves the user experience.

So please, if you’re reading this and you are the client, here’s a suggestion.

If you’re asked to test and submit changes, do that.  Say – I’ve worked through the app and here is a list of changes I’d like.  You can even list them by priority, or by – needs and wants.

If you think you have to give praise – say – thanks for all the work, I think we’re almost there, here are some final things…

Ok, I’m good now, got some changes to make.

I have a MacBook Pro and I have a ‘computer-all-my-life’ background and know the importance of backups and keeping them secure.  The institution I work at is still predominately windows – and therefore has no backup solution for mac computers.  I’ve been hounding them for months to allow me to purchase an external drive so I can use Time Machine to backup my system. This is my entire career of code here.  They dragged their feet and never came through.

The icon of Death

Two weeks ago the inevitable happened.  My drive crashed.  On a mac, when it can’t find the filesystem, you don’t get a geeky message like you do on windows explaining what is wrong.  What do you get? This precious icon to the right.  Apparently that is the universal icon for ‘you’re screwed’ with an Apple-esk design to it.  After much googling about this – it seemed that icon meant one of two things.

1. The problem could be simply the mac couldn’t find the startup disk, and you just reconfigure it using the iOS boot disk. Or – more likely
2. You’re screwed, and your filesystem is so messed up – you might want to go back to paper and pencil and use a real folder like the one in the icon for storage.

To make a long, depressing story short, my drive was toast, and I replaced it.  Out with the bad drive, in with the good drive.  We have an amazing disk image install process here – so all I had to do was plug in a network connection and power, hold down N and power it up.  The MacBook Pro then went out and found the image on our network and through a wizard – installed Lion (an upgrade even).

Data recovery efforts were grim.  I created a Knoppix CD and hooked the dead drive to an external SATA rig and attempted to get some specific code and MySQL databases off. These particular items were not in production anywhere, and were pretty important to me.  I was able to scavenge some code with no errors, but the databases I needed were all corrupt.  I’m guessing because they were opened when the drive hit the fan.  So I have code, but no database for it.  Yes, I can reconstruct the database from looking through the class files, determine the field names, etc … fun.

Enter Chrome.

Remember Chrome? That’s what this is all about  (we’ll come back to my database reconstruction efforts in a bit)  During the re-setup process of my laptop I had installed Chrome.  I prefer this browser over all others because its fast and doesn’t feel bloated like Firefox does (which used to be my favorite browser).  I now have a new appreciation for Chrome – because I learned something I had no idea about.  With a fresh iOS installation obviously I didn’t have a backup of anything, much less my bookmarks.  I open Chrome, sign into Gmail – and a little while later, I notice all of my bookmarks are there.  All organized like they were before the crash.  Not only that, but any site I had told to “Remember Me” for logging in – it did.  Very cool.  I even texted a friend and asked  – ‘How does chrome have all of my bookmarks!?’

I’ve since learned that Chrome is linked to your google account (or can be).  You can configure whatever you want stored by accessing Chrome Preferences .. or by entering chrome://settings/browser.  From here select Personal Stuff from the left, and you can connect / disconnect your google account

You can also choose to manage your synced data on Google Dashboard – this webpage (once you log in) allows you to stop the sync and delete the data that has been synced, along with a host of other google settings not related to Chrome.

Clicking advances settings gives you options to choose what you want synced.  From here you can also choose to encrypt all data that is sent – default is ‘encrypt passwords’ and change the passphrase – might be a good idea – depending how paranoid you are.

Database reconstruction

We’re back to my database woes.  Because I had the ever so controversial autofill checked to sync my forms, and since I originally hand built the MySQL database using PhpMyAdmin – when I got MAMP all set back up – I went in to build my database and tables.  Imagine my surprise when it did a drop down suggestion when I typed in the database name.    I created a new table and when I started entering field names again, it listed suggestions from when I had done it before.  Very cool.  So I was able to fudge my way through creating my tables again because I had some paper notes about the origianal names for things.

But what about the seed data I hand inserted using PhpMyAdmin?  Yep – that auto filled for me as well, the same data I had to think hard about when I put it in, just auto-suggested for me as I typed away names, and values.  So in a little less than an hour, I had my database and seed data setup and my salvaged code working.

Thank you Chrome and Google!

I got to thinking the other day how I was similar to a lawyer, but really should leverage more lawyer-ish practices.  How did this pop into my mind?  Quickbooks.  The software I love to hate.  At this point in my life, I use it for invoicing, managing client information and tax reports.  Its way more software than I need, and I got to thinking what other service industry might have software that does what I do.  I’m not aware of something (cost effective) that will do the basics that I want, granted I haven’t really done much research because I own Quickbooks (so feel free to tweet @thecodedog what you use for invoicing and book keeping).

When you set up Quickbooks, you choose a similar industry – and it goes off and sets up accounts and some of the ledgers, etc (stuff I don’t care about). Anyway I got to thinking how the functions of what I do are similar to a lawyer.  THEN I had an amazing idea that I think I’ll implement for 2012 – and we’ll see how it goes (I’ll get into that after these lists) –

Here’s how I’m similar –

• I AM AN EXPERT AT WHAT I DO
• I charge hourly for my expertise and work (unless its a bug fix)
• I take on projects, small and large
• I meet with potential clients to determine if I can help
• I advise people with my expertise (whether they follow it or not isn’t my issue)
• I manage multiple projects at once
• I research and learn all the time
• I create structured documents
• Occasionally I build and implement templates
• I reuse items from other projects
• I rarely produce a physical product
• success is determined by the client, peers and typically the public
• everything I do is to help the client
• I have a pretty damn good success rate

How I’m NOT like a lawyer

• I’m honest and not seedy (I know not all lawyers are bad)
• I’m 100% up front with pricing and charges and my clients ALWAYS know what to expect when an invoice comes
• I don’t charge for phone calls
• I don’t charge for quick tasks that take more effort to create the invoice than the actual work performed
• I probably do not charge enough (thus don’t overcharge) for the work I do
• I don’t have an assistant to do my dirty work (book keeping / phone calls, research, etc)

There is one thing that a lawyer does that is good that I don’t currently do, and it hit me and I’m curious if it will fly.  Keep me on retainer.  I looked it up and here’s a quick definition for a retainer –

A retainer is a fee paid to an attorney or other professional in advance, for services. Often, retainers are paid monthly, based on an estimate of the amount of work to be done for the client each month.

A retainer might work like this: You would pay your attorney $500 a month for legal services during that month. If you have a question or need a matter handled, it comes off the retainer amount. If you don’t use the full amount of the retainer, in most cases you would not get a refund or a credit toward the next month. If you use more time than the retainer amount will cover, you will need to pay the additional fees.

With that in mind, I’m wondering if any other programmer does this, and if clients would go for it.  In the past I’ve offered support contracts for an app I’ve built for a client.  If they wanted small changes, tweaks (all defined beforehand) they didn’t have to worry about it.  If a user had questions, needed training (again outlined specific hours) that was included.

The upside to a retainer agreement this is that you have reliable/consistent income each month, the client has the assurance you’re not going to walk away from the job.  They have the security in knowing that you’re there to call if anything comes up. They have a reliable budgeted amount to allocate for the project. And most importantly – if you have a retainer agreement with them, they would get preferential treatment and pop to the top of the workload (or as near as possible depending on what you have going on)

The downsides (two that I can think of) is that if you do this with too many clients, you might be overwhelmed with ‘high priority’ work (which would probably happen regardless of a retainer agreement and the same number of clients).  The other (seems petty) is that you probably have to invoice them every month instead of hoping they remember to pay you (which you may not have to do if you didn’t do any work for them that month).

Deciding I’m going to offer this…what’s a fair rate then?

Figuring here in Pennsylvania the typical rate for what I do is $70-$100/hr – are clients seriously going to pay me 4-5 hours ($280 – $500) a month so I’m at their beck and call?  I personally don’t have any clients that large – I think they’d say – no, that’s ok, you can just invoice us for work performed. Maybe if they do put me on retainer – I lower my rates? Even 50% seems a bit steep for some of them and then you’re just screwing yourself because now you’re performing 5 hours of work for half the price you would be if you weren’t on retainer for them.

Its an interesting issue I’ve been pondering.  I like the idea of offing this to clients, I guess if they take it, its a win either way.  Especially if I get a month in there where they don’t use the 5 hours.

If you do this or have done this – feel free to comment below or shoot a tweet to @thecodedog

and if you want me on retainer for you definitely contact me and we’ll talk – jonathan.wheat @ codedog.net 🙂