Archive for April 27, 2014

Sunday, April 27, 2014 [Tweets] [Favorites]

Car UI

Dr. Drang:

Apple’s recent announcements about CarPlay have revived the intermittent discussion on blogs and podcasts of the poor quality of user interface design in automobiles. Most of the talk has been about the phone and music player controls, but the UI problems go way beyond that. Yesterday, I came across a particularly bad example.

How Is Amazon Glacier Implemented?

Robin Harris argues, unconvincingly in my opinion:

Therefore, by a process of elimination, Glacier must be using optical disks. Not just any optical discs, but 3 layer Blu-ray discs.

Not single discs either, but something like the otherwise inexplicable Panasonic 12 disc cartridge shown at this year’s Creative Storage conference. That’s 1.2TB in a small, stable cartridge with RAID so a disc can fail and the data can still be read. And since the discs weigh ≈16 grams, 12 weigh 192g.

For several years I didn’t see how optical disk technology could survive without consumer support. But its use by major cloud services explains its continued existence.

sintaks (August 22, 2012):

Former S3 employee here. I was on my way out of the company just after the storage engineering work was completed, before they had finalized the API design and pricing structure, so my POV may be slightly out of date, but I will say this: they’re out to replace tape. No more custom build-outs with temperature-controlled rooms of tapes and robots and costly tech support.

[…]

I’m not sure how much detail I can go into, but I will say that they’ve contracted a major hardware manufacturer to create custom low-RPM (and therefore low-power) hard drives that can programmatically be spun down. These custom HDs are put in custom racks with custom logic boards all designed to be very low-power. The upper limit of how much I/O they can perform is surprisingly low - only so many drives can be spun up to full speed on a given rack. I’m not sure how they stripe their data, so the perceived throughput may be higher based on parallel retrievals across racks, but if they’re using the same erasure coding strategy that S3 uses, and writing those fragments sequentially, it doesn’t matter - you’ll still have to wait for the last usable fragment to be read.

skrause:

The author quickly dismisses hard drives because at the time of the Glacier launch SMR drives were to expensive because of the Thai flood. But after a few years of running S3 and EC2 Amazon must have tons of left-over hard drives which are now simply too old for a 24/7 service.

So what do you with those three year old 1 TB hard drives where the power-consumption-to-space ratio is not good enough anymore? Or can of course destroy them. Or you actually do build a disk drive robot, fill the disk with Glacier data, simply spin it down and store it away. Zero cost to buy the drives, zero cost for power-consumption. Then add a 3-4 hour retrieval delay to ensure that those old disk don’t have to spin up more than 6-8 at times a day anymore even in the worst case.

jeffers_hanging:

I worked in AWS. OP flatters AWS arguing that they take care to make money and assuming that they are developing advanced technologies. That’t not working as Amazon. Glacier is S3, with the added code to S3 that waits. That is all that needed to do. Second or third iteration could be something else. But this is what the glacier is now.

amznian:

I am an AWS engineer but note that I am not affiliated with Glacier. However James Hamilton did an absolutely amazing Principals of Amazon talk a couple of years ago going into some detail on this topic. Highly recommended viewing for Amazonians.

From what I remember from it, its custom HDs, custom racks, custom logic boards with custom power supplies. The system trades performance for durability and energy efficiency.

jpalomaki:

Having a robot juggling the hard drives would not make that much sense. The reason why we have optical disc and tape robots is that the tape and discs need a separate device that reads/writes them. With hardware there’s not such need.

With hard drives it would make more sense to do some development on the electronics side and build a system where lots of drives can be simultaneously connected to a small controller computer. All of the HD’s don’t need to be powered on or accessible all the time, the controller could turn on only few of them at a time. And of course also part of the controllers could be normally powered off, once all the harddrives connected to them are filled.

Reflecting on the New Look of National Park Service Maps

Tom Patterson (via Amit Patel):

This paper examines the techniques being developed by the U.S. National Park Service (NPS) Division of Publications for designing plan (2D) maps with a faux realistic look. The NPS produces tourist maps for 385 parks in a system spanning a large swath of the Earth’s surface from the Caribbean to Alaska to the South Pacific, and which is visited by nearly 300 million people each year. Many park visitors are inexperienced map readers and non-English speakers. In our ongoing effort to make NPS maps accessible to everyone, the design of NPS maps over time has become less abstract and increasingly realistic, particularly in the depiction of mountainous terrain and natural landscapes (Figure 1). Many of the techniques discussed herein are borrowed from or inspired by 3D mapping (Patterson, 1999). However, the scope of my paper deals exclusively with plan mapping—a format that has received scant attention in the digital era in regard to abstract vs. realistic depiction compared to the 3D world. It is also the format in which the majority of NPS maps will continue to be made.

Apple/Google Hiring Lawsuit, Settled

Dan Levine:

Four major tech companies including Apple and Google have agreed to pay a total of $324 million to settle a lawsuit accusing them of conspiring to hold down salaries in Silicon Valley, sources familiar with the deal said, just weeks before a high profile trial had been scheduled to begin

Tech workers filed a class action lawsuit against Apple Inc, Google Inc, Intel Inc and Adobe Systems Inc in 2011, alleging they conspired to refrain from soliciting one another’s employees in order to avert a salary war. They planned to ask for $3 billion in damages at trial, according to court filings. That could have tripled to $9 billion under antitrust law.

Joe Mullin:

There were more than 60,000 workers in the class. Class members claimed that the “no cold calls” agreement resulted in $3 billion of lost wages, a far cry from the settlement agreement.

The New York Times:

The companies, which are some of the world’s richest, must think that is a bargain. At a moment when Silicon Valley is losing some of its luster even on its home territory, the antitrust case depicted the upper levels of the valley’s executive suites as a cozy old boys’ network. Private deals are made, and then the executives send emails saying they wanted everything to remain secret.

[…]

Originally there were seven defendants. Settlements with Lucasfilm and Pixar (both now owned by Disney) and Intuit were reached last year. Those companies agreed to pay a total of $20 million — small change in the valley.

Sounds like clear victory for the defendants. They avoid more embarrassing e-mails and testimony and end up paying just a few thousand dollars per employee, surely less than they saved through this scheme, which also suppressed the wages for plenty of other employees outside the class.

OmniFocus 2’s Low Information Density

OmniFocus 2 for Mac adds some cool features. The inspector sidebar and Quick Open seem especially nice. But it also includes a new layout that I think is a regression. I’ve been worried since last winter that OmniFocus 2 would ship with an iPad-style two-line design. It seems cluttered with icons and context names that get in the way when I’m scanning, while showing much less actual information in the same amount of space. (OmniOutliner 4 also reduced its data density, but it included options in the Pro version to tighten up the spacing.) The current OmniFocus beta, which is quite a shock to an OmniFocus 1 user, is actually the new and improved design:

In today’s most recent build (r207056), we’ve significantly reduced the amount of vertical whitespace in the main outline for actions and projects. OmniFocus 2 can now displays 65 rows in the same amount of space as it would previously use to display 48 rows—an increase of over 35%.

Ken Case says that Omni is listening, but the layout’s been this way since the first public screenshots—and presumably long before that internally. This is obviously not a high priority. The user interface is now frozen for the June release.

The new design also has the checkboxes on the right, like in iOS, but I don’t think this works as well with a wider window. The left sidebar with the project and context has a minimum width that’s about twice what it should be, wasting further space. They’ve also removed the features for customizing the fonts and styles and reduced the filtering and sorting options.

I consider OmniFocus 1.x to be one the best Mac apps ever, so it pains me to see it seemingly ruined by iOSification. It looks nice, but I don’t think it works well. Obviously, the developers are no dummies. Maybe they are right and this is what most people want.

Nevertheless, it puts fans of the old version in a bad position. Soon 1.x will no longer be supported. Someday it will stop syncing with the current iPhone version, or break in some other way. There’s no telling when or even if 2.x will match it. There are lots of competing apps, but I haven’t found any of them compelling.

Update (2014-05-20): Ken Case announces a work in progress:

If you’d prefer to see all of your task information laid out in one line (so it’s more vertically compact) and would prefer your status circles on the left, you can start experimenting with this now by opening this URL[…]