stephbu

Yeah me too in 1982, using the Melbourne House Z80 reference, aged a young 10 years old. Working with POKE and no macro-assembler, I wrote mnemonics then translated them to machine-code by hand. A baptism of fire that to this day that I've not forgotten.

This book was the ignition that changed my life... https://archive.org/details/z-80-reference-guide-alan-tullya...

Ironically the power per cycle is decreasing - power and thermal dissipation are really the limits NVIDIA is exploring. It’s what the software does with those cycles that is leaping exponentially.

You're right, it's astroturfing a placeholder in the market in the absence of product. The difference is probably just the target audience - feels like this one is more aimed at share-holders and internal politics.

Landlords -> lobbyists of large real estate investment corporations. One of the biggest social crimes is enabling housing ownership companies to become large investment vehicles.

Post-production QA is an industry problem in general - the difference between systems is who does the QA.

In the dealership model - after transit, they do a once over and post-production/transit repairs before the car appears on the lot. So much so that most states wrote legislation to limit repairs allowed while still being called "new". (often ~5% of the retail value of the car). Note that dealers also pay wholesale rates on the cost estimate for that, so this can be quite large repairs.

The T* direct-model should put the SC in that same spot. Bug is they're not doing the work - probably the emphasis on throughput incentivizes the wrong behavior. All too often the the SC tries to palm-off problems and/or the consumer has to do drive QA. Fixing it means the customer is on point - obviously YMMV.

IMHO this is a huge gap and flaw. Unfortunately OEM dealership behavior is so predatorily atrocious that even this flaw isn't enough to overcome the otherwise positive T* sales experience. I now know how few signatures are needed to buy a car in my state. I have zero intention of participating in the "sit outside of a finance office for an hour, no I don't need Scotchgard, interest rate manipulation" routine ever again.

Dealerships often talk about "relationships" - this BS sales talk - we're just prey to them.

Doesn't matter if you like or dislike the T-word or EM-himself. Fact is they have built one hell of an experienced/industrialized muscle in this field. That experience is priceless right now - Ford, VW, GM etc. don't just need to run-as-fast, they need to run faster. All the while not repeating the mistakes that others were allowed to make when there were less optics on the problem. This technology change is going to be a capital-intensive, painful experience for the incumbents.

Datacenter mode - ~11yrs in the making - nice find

“Remove my details” that asks that I sign-in to Google. Privacy-washing.

Consider what the Union goal is - expansion of membership and increased funding/power/leverage. Uber’s goal is brand-washing. Fighting employment law is a cynical win-win for the Union and Uber in this instance - member welfare be damned. Tells you a lot about the intentions of both the union and employers.

Funny thing is this cycle seems broken. So many deep pocketed competitors have stomped in and skewed the content market by showering money on all the producers.

In turn this has drove up the pricing, enabled exploitation by subpar producers, and over-extension of streamer capital in a bid to stay competitive.

Implosion and consolidate seems inevitable, the market just isn’t big enough to support 5x Netflix sized companies.

I spend more time on Netflix looking for things to watch than watching things.

This isn't necessarily a Netflix thing - it's a streaming industry problem - driven by short supply and high demand with having multiple competitors in the market with big bank balances. When Netflix first started, they were pure quality distribution, now they and other streamers are bleeding money in bidding wars for content generated primarily by lame "chose one word from each column" production houses. Aliens, Teenage, Investigators - out pops humdrum 8 episodes of binge-fodder.

I went to streaming to escape from the cable playbook. I anticipate that all these additional revenue tactics will result in me dropping Netflix, and others. Maybe I'll return once the market has consolidated a little. Maybe not.

At 550km altitude, each Starlink satellite in low-earth orbit has a visible horizon of only about 700mi, and I suspect usable range that is much smaller, probably low 100’s of miles. To extend range to a ground-station beyond that will probably take multiple peer satellite hops - I suspect that inter-satellite bandwidth is a a precious commodity - and priced as such.

Seems unlikely, more probable is that you’ll have other problems. I suspect there are other differences in the equipment and service delivery to tolerate ocean conditions.

Maybe in 1,700yrs they'll see us...

Used magnet for years - absolutely gamechanging for OS/X when coming from Windows/snap-window

Came here to write the same - that was amazing...

I spent a couple of decades in big-tech, and of-that did adtech for about 5yrs - we captured, aggregated, cooked out activity streams from billions of toolbar, browser, and beacon events per day, for real-time, long and short-term user profiles for around a couple of hundred million people - primarily for feeding the ad-exchange behavioral targeted ads. Back then about 80% of the active internet users. This was a huge business even back then - billions.

Since then the world and his wife have captured data and inserted telemetry beacons everywhere. Consider how many services you use that are today subsidized by “anonymized” selling/trading/sharing/merging user event streams - the TV your watching, your cellphone provider selling user behaviour data, even your ISP is selling your DNS lookups - they’re all at it now. Worse, mobile has made it much easier to install platform frameworks that offer developer features in return for data-collection such as location, user profile etc.

I don’t think it’s overblown - that “anonymity” isn’t that anonymous when you add enough dimensions - you just haven’t seen how massive the data broker business is.

Ads is just one of their businesses, consider FBLogin and 3rd party data-access e.g. full-name, date-of-birth, email address, friends, the list goes on...

Pretty much every cellphone/app user dimension is available for a price on the open market. What you do think Facebook et.al, sells when an app user allows background tracking?

Is it unethical to study behavior of the populous? If anything bringing it into a clinical study probably brought more oversight in terms of ethical handling, and aggregation/anonymization of data than the source would provide.

Headline feels a little misleading, they didn’t track phones per se, they bought data from one of the dozens of cellphone data-brokers that continue to operate despite legislation and congressional action/in-action.

At what point are "eyewitness testimonies" and "identity parades" going to legitimately be called out as questionable. The science certainly points towards their lack of credibility in many many circumstances.

I guess there is no real incentive on the part of the criminal justice system to resolve this.

They already do - Residential Energy Code - 402.1.1

TLDR - R21 in the walls, R49 in the roof, R30 in the floors.

It’s paired with credits for retrofitting to a reasonable level.

This is probably the most insightful answer, the system is geared to rewarding creation of the next greatest technology. Sustaining what ever your predecessor did just doesn’t get prizes.

It feels like Google’s attention span, leadership longevity, and product development patience is roughly 3 years. Any product that survives longer than that probably has transcended beyond being a “pet project/toy” into a PR-problem or revenue-stream significant enough that it takes on a life of it’s own. As management turns over, that lease on life is renewed…

https://killedbygoogle.com/

While this fosters new ideas and opportunities, that conversion into long-term direction and execution can be pretty rough.

I think I covered those poin in calling out the cost model elements - renting the “bottom third” of the costs via managed facilities, BMaaS etc. helps in terms of reducing or eliminating capital expenses and some human toil especially in a more stable business that doesn’t benefit from per-minute lease terms or multi-year. As I called out elsewhere, sizing the hosting model to the economic model is really important.

Even in the pure rental or managed BMaaS, the human cost can quickly dominate the economic model. Owning machines and OS’s is expensive at anything more than a couple of racks of machines. Eliminating people and human change/release from touching things in the datacenter is probably the first priority. Otherwise it is hard to consistently drive that human number down and meet service quality expectations for 24x7.

Fair point, I guess the point is that unallocated resources - space/power/servers etc. can become huge stealth money sinks, eating budget every hour of every day. Being cognizant of the consumption economics before you stump up for resources is important, as is fitting the investment model to match those economics. Setting utilization/allocation targets are just one way of measuring if those models efficiently match. This is true for any service or resource consumed.

I broadly agree with the statement that owning an OS in general is toil that applies any of these IaaS/VPS/BM scenarios.

Owning a BM server different toil - server parts fail, the network it attaches to needs control and it fails too, firmware needs updating more regularly than ever, DC space needs managing over time etc. For a small number of machines maybe this is NBD. For thousands of machines this is just grunt work which while automated, still needs change management and control - rebooting the whole fleet in the middle of the day definitely opens doors in your career.

Doing everything you did in the DC in the Cloud is absolutely the worst way to adopt Cloud. Owning an OS is a non-goal, you’ve gotta climb to a higher abstraction - workloads, and quit caring about machines. This is where most companies fail.

It’s pretty hard to generalize this without qualifiers. Electricity can be the most expensive problem, but it requires carefully planned control of the other factors - e.g large scale with highly automated servers, network, and meat-reducing control planes to become true. Otherwise factors such as people and under-utilization can especially dominate smaller and/or less efficient facilities.

Thinking thru the factors, many seem obvious, but are often forgotten/ignored when comparing rental or IaaS costs.:

Space is a fixed cost driven by market rates and maximum Server Capital Costs i.e. floor space. Failing to fill the room increases your cost/server efficiency. Pretty common to run out of thermal/power before you run out space, as equipment efficiency increases through the lifespan of the facility.

Server and Power costs scale together, carry a minimum cost for keeping machines on, and vary based on utilization. Again if the servers aren’t doing work, your efficiency ratio will drop. Larger space typically have pre negotiated power commitments too - failing to consume that carries fiscal penalties. Servers unit costs are fairly cheap, storage not so much. Full utilization throughout capital/lease lifespan is the goal - anything less increases relative cost/server.

Network costs scale with Server Costs, and vary again by utilization - minimum invest rules apply, all servers need at least one network port, as well as upstream Core/TOR/Miniswitch gear. The network gear lifespan is typically longer than servers, but shorter than facilities. It usually incurs annual support/maintenance charges too. Bandwidth charges are variable as expected.

People costs scale with a step-function and numbers driven by minimum coverage requirements, task complexities, and level of human toil. Performing any task on a device by-hand is expensive in most markets - touches on tickets, change management, task time etc. Fully burdened S+R in Western cultures is typically ~2x the salary - a $80K employee probably costs around $150K by the time all the workplace costs, taxes, and benefits are paid. Network folk are typically premium resources compared to DC Ops. Sustainable 24x7 coverage looks like a staff of 3-4 people.

To add to this statement - don’t make any investment unless you have a solid plan to fully utilize it for the full length of the purchase terms. This applies as much to AWS @ minute terms, as it does to a rental @ daily terms, as it does to co-lo capital investment @ yearly terms.

Watching the “Making of the Interstellar Soundtrack” is quite simply inspiring. The process and mechanics of making such moving art with such primative tools is amazing.

“On the Nature of Daylight” by Max Richter used in the opening scenes of Arrival is one that really punches you in the gut - if the theme and monologue isn’t enough to already give you a lump in the throat as a parent.