Ars Technica has a feature on the latest developments in real-time applications on the web. Filed for near-future development…
Bob Cringely outs IBMs plan to slash its North American workforce by 85 percent… and no one cares. Or at least, his series hasn’t generated any interest from the business press, who are probably waiting for IBM to hand them a release rather than waste valuable time going in to dig up internal memos and such.
Why does it matter? Because for the last two decades (and more, of course!) IBM has been selling multi-million-dollar business systems to anyone who can afford them. Including your local government, hospitals, police, firefighters, utility companies, and so on. Really big, expensive, complex systems that have changed the way that cities do things. (Or so we’re told in the ads.)
Those systems are big and complex for a reason. They didn’t have to be, but you don’t get million dollar contracts by proposing simple solutions. You also don’t get decades of guaranteed support revenue by building something the client can manage themselves. Like it or not (and I hate it!) administrators have spent countless amounts of tax dollars wedding their operations to IBM systems.
So when memos leak that promise to completely gut the company, wouldn’t you think that was news? Are we going to wake up in two years to a crumbling software infrastructure, with no money left in the budget for new construction and no vendor left to honor the warranty? Probably.
Fortunately, if Cringely is right, there will be a lot of IBM-trained programming talent looking for work by next year.
CrashPlan is a neat bit of freemium-ware that allows you to easily and securely backup files on one computer to another.
If the destination computer is on the other side of the internet or a slow wi-fi connection, it makes sense to seed that backup by first backing up locally, and then transferring that archive to the remote computer by hand (aka sneakernet).
I tried to set this up using the instructions here, but I kept getting the error “incorrect archive: belongs to another computer”. Perhaps they apply to an older version of CrashPlan, I don’t know.
The following assumes you have CrashPlan client installed and logged into the same account on both local and remote computers, and that remote is accepting inbound backups.
Step 1: Create a local archive using the local computer.
Create a local folder destination and perform your backup. Within the local folder there will be a folder with a long numeric name (like 517753233945951089). This is the archive folder.
Step 2: Create a remote archive using local computer.
On the local computer, go to Destinations > Computers, select the remote computer and click Start Backup.
After a few seconds, pause the backup so you don’t waste bandwidth copying files. All you’re doing is getting the connection set up.
Step 3: Take/copy the local archive (henceforth the copied archive) to the remote computer.
Step 4: On the remote computer, switch the inbound backup archive location to the copied archive.
Go to Backup then click on the inbound computer name to see the Inbound backup settings.
Click the folder icon to the right of the Location to change it from the default location to the location of the copied archive
CrashPlan will ask you if you want to delete the incomplete archive from the default location, and you probably do.
Step 5: Unpause the backup on the local computer and verify that it is now magically complete! Congratulations, you have seeded the backup successfully.
Step 6: You can now delete the original local folder destination on the local computer, since you are now backing up to the remote computer.
This will be famous every four years:
When the GA creates the transfer certificate, it gives it a one year validity range. It uses midnight UST of the current day as the valid-from date and one year from that date as the valid-to date. The leap day bug is that the GA calculated the valid-to date by simply taking the current date and adding one to its year. That meant that any GA that tried to create a transfer certificate on leap day set a valid-to date of February 29, 2013, an invalid date that caused the certificate creation to fail.
From Summary of Windows Azure Service Disruption on Feb 29, 2012.
First half of a 1982 interview with Rear Admiral Grace Hopper, inventor of COBOL and free radical behind the computerization of the Navy. Greatest explanation of a nanosecond, ever.
Second half: https://www.youtube.com/watch?v=CVMhPVInxoE
Google sync on Windows Phone 7.5 works great, with one notable exception: your shared calendars aren’t automatically recognized.
Fortunately this problem is solvable on Google’s end, as detailed in this post. Basically, you tell Google to sync specific shared calendars with your WP7 device, and you’ll be all set. Worked for me.
The bitter family feud that has divided the children of Martin Luther King Jr. isn’t much different than other fights between brothers and sisters — except that this one has spilled into the courts and publicly tarnished the legacy of an American icon of peace and harmony.
“I have a dream that my four little children will one day live in a nation where they will not be judged by the color of their skin but by the content of their character.”
So that dream came true, just not the way he expected.
The President recently asked us to come up with a better system for enforcing copyright online.
Here’s my proposal:
Exempt linking to, copying, and redistributing digital information, while leaving all other copyright protections in place.
This radical but progressive move acknowledges the fact that digital copying and redistribution is fundamental to the way networked computers operate. It protects all manner of fair use sharing and derivative works as long as they happen online. And it frees up resources for enforcement of copyright on physical media and the sale of licenses, which is where media companies make their money.
In short, everybody wins. And yes, people will still buy books, movies, and music when copying them is free. We are willing to pay for convenience and authenticity, and of course we want to own nicely packaged copies of the works that we treasure.
A file that big?
It might be very useful!
But now it is gone.
In 1998, Salon Magazine ran a challenge to readers to convert Window’s style error messages into haiku. The above, by David J. Liszewski, is a favorite. But the winner in my opinion, was this gem by Nick Sweeney:
wind catches lily
scatt’ring petals to the wind:
segmentation fault
Facebook-er Kent Beck’s Software Design Glossary includes an entry on Succession, “the art of taking a single conceptual change, breaking it into safe steps, and then finding an order for those steps that optimizes safety, feedback, and efficiency.”
As an example, he gives us (for free!) this pattern for migrating from one datastore to another:
- Convert data fetching and mutating to a DataType, an abstraction that hides where the data is stored.
- Modify the DataType to begin writing the data to the new store as well as the old store.
- Bulk migrate existing data.
- Modify the DataType to read from both stores, checking that the same data is fetched and logging any differences.
- When the results match closely enough, return data from the new store and eliminate the old store.
Beats shutting everything down for hours while you wait for the data to get copied over the wire.
Re:Part of a money conflict within the King family (Score:5, Insightful)