Posts by Fazal Majid

Property taxes go to your city, not the State of California

Minor nit. The State does cap the property tax, but it goes to local authorities.

Hubris

Building a transport protocol on top of UDP is usually a red flag. The 2RTT TCP handshake is there for a reason. There have been proposals to simplify it, e.g. T/TCP, but those have not been widely deployed, nor have extensions to TCP like SACK (RFC 2018) or Quick-start (RFC 4782) that address other shortcomings in TCP.

Many of the proposed "fixes" to TCP over the years were designed by people who do not understand the issues, work only in best-case scenarios and lead to severe failure modes or even affect the stability of the network. Considering all the proponents of QUIC are browser engineers and there has been little to no review by protocol design specialists like Sally Floyd, I would err towards assuming this is an illustration of the Dunning-Kruger effect at work and yet another half-baked idea from Google with the potential to do real damage.

Not surprising

Justified, even in light of the Snowden revelations, even if the primary motivation is probably old-fashioned protectionism.

This should be compared to the Amazon FireTV Stick

Which only costs $40, for equivalent hardware. Gives some idea of how much Amazon is subsidizing them.

iCloud Storage

My guess is they realized the current backend for iCloud Storage won't scale, want a replacement and don't want to be beholden to a third-party software vendor for something this critical to their platform. iCloud Storage has 2 APIs, document database and key-value store, and FoundationDB's multi-model capabilities would be a good fit.

Fuel cells may be dumb for cars because of the bulk and weight, but they are a pretty good fit for solar electricity storage.

There's also the Quantcast File System, a C/C++ drop-in replacement for the Hadoop file system.

In other words they stuck some Illumos based distribution like OpenIndiana or SmartOS on top of commodity hardware, added a Web UI, and called it a day. ZFS deduplication is a notorious memory hog and not recommended for production use. Compression works very well, on the other hand.

As for BTRFS, it is at least 5 years behind in terms of maturity and robustness, ZFS is available for Linux as well, just not shipping together because it's license is not ideologically pure enough for the ayatollahs of GPL.

Good summary, here is some extra color

My credentials: I worked for France Telecom's R&D, hold one patent on online micropayment systems and deployed a successful one in the early days of the Internet (1997 or so).

There are 2 prizes they were all contending for: the ability to skim commissions from all mobile payments, and the ability to mine customer transaction data.

The banks (well, credit card issuers) own both in the current credit system. They have historically been extremely paranoid about telcos, and successful m-payment initiatives like M-PESA in Kenya (with which Vodafone essentially disintermediated the banks and corrupt government) certainly stoked that.

Google obviously hungers for #2, #1 being mere cherry on the icing.

The handset makers had a much weaker position, as in the US at least the telcos are their biggest customers and hold the whip hand, as demonstrated by Nokia's failure to penetrate the US market to the same extent as worldwide.

Apple never entered the fray until Apple Pay, unlike some of the Android handset makers. Apple Pay essentially implements banking industry standards on the banks' terms, for a very modest commission. In other words, they cut a deal with the banks to get an acceptable #1 and forgo #2, and in the process gave the banks a way to snatch victory from the jaws of defeat at the hands of Google or the carriers.

In the West, iPhone users are the choice demographic in terms of spending power, and I expect Google will have to cave to the banks' demands and implement a form of EMV tokenization similar to Apple's. It's unclear whether they will be allowed to ingest transaction data. The carriers have already bowed to the inevitable.

Big Data is overrated

Big is not the same as valuable. Just because you know your customers does not always mean you understand them or can manipulate them, in other words it is not always actionable. Tesco's meltdown, despite their loyalty card prowess, and the fact they are willing to ditch the alleged Big Data Crown Jewels, indicates they now agree.

CBC mode is not considered best practice any more

I suppose it could have been worse, as they could have used ECB, as a surprising number of "security" products still do.

The keypad is OS-independent, which is nice for Linux users, or Mac users tired of waiting for IronKey to update their unlocker app for the newest version of OS X.

It's nice they used a solid aluminium chassis for the drive. Boo for USB2. One of the great things about the IronKey S200 is that it uses SLC memory, which made it a very speedy drive (unlike the horrendously slow MLC IronKey D200). I wonder where the Toshiba fits, performance-wise.

I have always disabled Flash entirely on my primary locked-down browser (Chrome), but the last incident made me reach my tipping point. My plan is to remove Flash entirely from my Mac, and leave it in a VirtualBox VM ghetto for when I absolutely need it. That way I won't have to restart all my browsers each time there is a security update, and the damage from compromise is contained.

The flaw in this plan is that Chrome bundles Flash, so there would still be the taint of Flash on the main OS X.

Oddly enough

France Telecom never tried to use the Orange brand in its short-lived foray in the Netherlands, a country actually ruled by the House of Orange-Nassau (confusingly, the eponymous Orange is actually the city in the South of France).

FT is phasing out its branding for Orange, which is basically the new, country-neutral name for FT both in France and worldwide.

Theory and practice

Typical academic paper making simplistic and very optimistic assumptions about failure modes. In my experience about one third to one half of storage faults are Byzantine, I.e. the drive doesn't just go down, it is actively attempting to sabotage your array by sending interfere down the bus (specially on buses where this is theoretically impossible like FC-AL) or all sorts of crippling behavior. Something like that will still require physical intervention.

Here is an excellent introduction to the subject:

http://dtrace.org/blogs/wesolows/2014/02/20/on-disk-failure/

And of course John Gall's immortal classics about systems thinking.

4K is now

I bought a house 6 months ago. It came with a media room with a 20 year old 42" 720p LG plasma with massive amounts of burn-in and frizzy pixels. My 85lb 46" Toshiba was heavier than the 76lb LG and I wasn't sure the wall mount was rated for the weight, so I replaced both with a 50-inch Samsung HU8550 that weighs half their weight, yet barely takes more space due to the thinner bezel.

The premium for a 4K screen is under $250 over comparable 1080p TVs, so it's a no-brainer for an appliance that is just too inconvenient to install more than once a decade because of its bulk. I went over reviews for the handful of models certified by Netflix for 4K streaming, so I do have some content available via the SmartTV, including Amazon now (no AppleTV or Roku or whatever supports HDMI 2.0 4K resolutions).

Paucity of 4K is not an issue, as I always expected the primary source would be my photo collection (4K=8MP) and 4K home video should become mainstream within a year or two as hardware H.265 compression chips find their way into smartphones. The Samsung doesn't have good photo casting options for a Mac user (surprise surprise), so I have to use a USB stick, but it works and is gorgeous.

I would say buying a 2K set today would be just as short-sighted and penny-wise-pound-foolish as buying a 720p set 5 years ago.

He is stunningly Naive

If he thinks the threat model for investigating high-level corporate malfeasance should not defend against state-level actors. All evidence shows the NSA has a sideline in economic espionage, whether from deliberate policy, horse-trading for reciprocal favors or simply personal corruption of NSA leaders is irrelevant. It is highly likely big establishment firms like BP or Unilever benefit from the same chumminess from GCHQ, and so on.

Last hurrah

You've got to wonder how much of this is driven by the transition from tape to disk. Certainly primary storage on laptops and enterprise first-tier is going all-flash, and mobile was always thus. The HDD iPod was discontinued. Hard drives seem to be increasingly relegated to cold storage.

Joyent's Solaris-derived SmartOS shows how containers (a.k.a zones) can coexist with KVM-based VMs on the same kernel. All modern Linux distros have similar capabilities, if not quite as refined. The battle is about management tools - the company that controls the de-facto standard can make a lot of money, see how VMware gave away ESXi, the real revenue is in vCenter, and value-added features like HA and vMotion.

Both public cloud and hypervisor vendors will gain container capabilities. AWS and Microsoft have already made announcements, the others, including VMware, will follow. It seems to me new applications will be designed for, and run directly in containers, whereas heavyweight VMs will be reserved for migrated legacy workloads. Containers do require automation tools like Puppet/Chef/Ansible/Saltstack to be manageable, however, as does the Cloud. Another opportunity to sell to the enterprise.

The efficiency gains from containers are nothing to sneeze at, you can squeeze an order of magnitude or two more containers than VMs on the same hardware, not a mere 10%. For cloud providers, specially PaaS ones, this is compelling. Even for IaaS, thin provisioning is easier to achieve with containers. Linux based container solutions need to reach the levels of maturity of Solaris, specially as concerns security as the recent Docker vulnerability shows. Using a better file system like ZFS (as done by SmartOS or Flocker) is also a big boost, and can provide something close to vMotion in terms of ability to migrate workloads, if not yet online (shutdown required).

Some of the more important gains are in the realm of latency - SSDs give, and VMs take away. At my company, switching from AWS to a containerized private cloud (OpenIndiana) yielded significant improvements in cost (6x), latency (1/3), throughput (3x) and uptime (MTBF went up 30x).

I've already stated my belief VMware style hypervisors will be relegated to a niche of hosting legacy workloads. Nothing wrong with that, and it can be quite lucrative, as shown by IBM. Container vendors won't be able to extract the same profit margins, because they are built on open-source, so the legacy vendors may still end up gobbling up the new entrants. In other words, legacy workloads may represent a small fraction of future volume, but a large portion of value.

Lots of fat to cut from their budget

Mozilla suffers from a sort of Parkinson's law - make-work expands to fill all available budgets. They have many utterly useless projects like Open Badges (http://openbadges.org/) that could be axed completely with no effect whatsoever on their mission (if anything, getting rid of the distractions should improve the focus on what matters, the browser and the mobile OS).

Damning Facts

IIRC John Lettice is the Editor of The Register, and his words carry weight.

Emil Michael mentioned specific budget and manpower, in the context of a sleazy PR event, and his previous job was at Klout, the epitome of social-media douchebaggery. This was not an off-the-cuff remark, but a trial balloon for something that has already undergone a feasibility study. His incredibly vague job title (VP of Business? Really?) suggests skulduggery is his real job description and all the flak Uber has been receiving is if anything understated.

Yeah, right

AT&T's fiber plans are vaporware, and have been for the last 20 years, despite getting a big chunk of the $200Bn that were supposed to support fiber rollouts.

http://www.pbs.org/cringely/pulpit/2007/pulpit_20070810_002683.html

Verizon did have the good grace to invest in its infrastructure (FiOS), although it has frozen further rollouts, but AT&T/SBC's corporate culture is to milk its rotting infrastructure for profits and never invest back into it.

A rose by any other name...

It's fairly straightforward. "Software-defined storage" is a meaningless phrase coined up by storage marketing shills in an attempt to capture the halo effect that currently surrounds "Software Defined Networking" (SDN), which is still going strong in the hype cycle, with its positive associations of cost savings by commoditization.

Storage, like networking, is one of the few areas of the IT industry that has managed to preserve its fat profit margins, but distributed storage architectures like Hadoop from the Web-scale world are putting an end to that, and the high-performance storage is going direct-attached SSD anyways, as the latency of a SAN or NAS array is unacceptably orders of magnitude higher.

It's not as if WiFi ever gets remotely close to the maximum theoretical bit rate, even with 802.11ac, so saturating Gigabit Ethernet is still a distant pipe dream.

Apple could ban the CurrenC app in tit for tat, but they probably won't bother as it looks likely to be stillborn. The Wal-Mart demographic doesn't overlap Apple's too much, but CVS should definitely be concerned at losing market share to Walgreens.

US chocolate stereotypes are 20 years out of date

European expat here, been living in San Francisco for nearly 15 years. Comments about American chocolate are ignorant. Just because LA is a wasteland doesn't mean Hershey's is the only choice available. There are some world-class US-based chocolate makers like Amano (Salt Lake City) or Guittard (Bay Area) that can compete with the best Europe has to offer. The US artisanal chocolate scene is vibrant, as is the bean-to-bar movement. Remember, the US is a huge and wealthy country, and even if the average standard of chocolate is abysmal, a small fraction of connoisseurs can easily sustain quality suppliers.

US chocolate standards are stricter than Europe's, as only a product made with 100% cocoa butter can be labeled as chocolate, whereas in Europe, because of British lobbying "chocolate" can be legally adulterated with up to 5% margarine. Granted, Hershey's is lobbying to water standards down to European standards, but they haven't succeeded yet.

Minor improvements

Apple hyping this release as the biggest ever is ridiculous, as the user-visible improvements are minor. There are 2 I was eagerly awaiting, however: Duckduckgo as default search engine, and reporting which apps are battery hogs.

"Google-powered devices will be equally attractive to those who value their privacy."

Google is the #1 threat to privacy, even ahead of the NSA. No one who values their privacy uses their services. The single greatest feature in iOS 8 is DuckDuckGo as a search engine option.

Yay for 4K

I just bought a 50-inch Samsung HU8550 for $1800 (after $700 instant rebate) to replace the ancient 42-inch analog 720p plasma set that came with my new house (burn-in and crazy pixels galore). This unit is certified Netflix 4K compatible, but I fully expect the primary 4K content will be from projecting photos. A 4K TV has 8 megapixels and is ideal for that purpose. The price is about double what an equivalent 1080p unit costs today, and most likely the price will fall down to the current level within a year or two, at which point no one will buy a 1080p model, just like no one buys SD or 720p today.

The key is to buy a set with HDMI 2.0, HDCP 2.2 and HEVC/H.265, which only became available in 2014. To qualify for the UHDTV 4K label, TVs also need 10-bit color, which is not yet widespread.

Ghafoor has known about this for a while, and has been at the forefront of exposing warrantless surveillance, hence the retaliation. Here is his lawyer six years ago recounting the Kafkaesque process of suing the government for it:

http://www.salon.com/2008/07/09/alharamain_lawsuit/

I just tried it

It's like the Netflix of music. Exceedingly limited collection, even for music older than 6 months.