osolo

I don't think they've announced publicly what models we're using. I don't think there's any particular reason for this, but just in case, I can't name them here. I'll see if this can be addressed in a blog post of something.

I can tell you that Recall isn't using Phi. Rather, it's using a collection of models that are much more tuned (and therefore much more efficient) for the feature.

My team worked on this feature and I can confirm that yes, semantic search is being done for both text and images. This allows you to do fuzzy searches like you mentioned in your comment and use words to match images. Everything runs on your device (all the models, the vector database, etc.) in order to preserve privacy.

I wouldn’t do that. Let me explain this from the perspective of a hiring manager. For a compelling position, it’s not unheard of to get 100 resumes per one spot (the ratio for my positions about 1:85). That’s a lot of resumes to go through. Often you’re hiring multiple people, so there are literally hundreds of resumes!

You can’t possibly interview everyone who applied, so you have to filter. Sometimes it’s easy (eg you’re hiring a senior but the candidate has only 2 yrs experience) but that’s rare. You have to develop more filters, or you’ll drown.

If I see a resume with a recent gap, it presents a question: What’s up? Why was this person not employed? There are many reasons, many not good. So I’ll pass, there are many other resumes without this problem.

Doing your own thing doesn’t mean there needs to be a gap. Write what you did! Call it a Startup or Startup Explorations. Talk about the idea, the tech, the UX, the marketing, all of it. This will show that you’re a self starter, a problem solver. Leaving the years empty doesn’t show any of that.

Let me leave you with this pragmatic thought: We can argue here if chronological resumes are good or bad, and what we should have instead. In the real world, that doesn’t matter because everybody else is submitting a resume, and you want yours to stand out for the better, not for worse.

The attitude in these threads always boggles my mind. When YOU, your teammates or someone else you know makes a mistake, then it’s just a mistake. But when someone at Microsoft makes a mistake it’s because Microsoft is evil and by extension so is the person who offended.

“It’s a big company”, you say. They should have a process! And I’m sad to say, they do/will have a process exactly because of this attitude. Then, everyone wonders why it takes forever to fix that bug they’re furious about or why features are slow to come.

Full disclosure: I work for MS, though have nothing to do with this. It’s a huge company. I know hundreds of people and most of them are upstanding and try to do the right thing.

My kids complained today that Google Classroom isn't working. After a quick investigation, I noticed that Snort on my firewall blocked the relevant Google server due to incoming TCP port scans. Sigh.

Imagine you're Expedia, and you want to make this consumer friendly change.

Now, on your website, all the hotels seem to be more expensive.

Nobody will understand the intricacies of why, and they'll simply book somewhere else where hotels are "cheaper".

We need regulation so these fees appear industry-wide.

$3700/2.5 years in electricity vs projected $12,500 in gas. So about $8800 in savings over 2.5 years, ~$293/mo, ~$3500/year. Still makes it hard to justify the Tesla based on just the gas savings, unless he keeps the car for >15 years.

Thanks for sharing this. It's exactly what I was looking for.

tldr; It clearly shows that HGST has an overall superior survival rate over time. WD is a distant second and Seagate in third (although the Seagate ST4000DM000 model is exceptional and fairs very well).

I really appreciate these blog posts and have used the data they present when buying HDDs for personal use.

One thing that bothers me is that the data presented doesn't really take into account the age of the HDDs. For example, if a batch of HDDs of a particular model is 6 years old and has a failure rate of 12%, that really doesn't tell me much except that it's an old HDD.

What I'd like to know, for a given model, what the blended failure rate is after 3mo, 6mo, 12mo, 24mo etc of operational time. That would be a real apples-to-apples comparison.