Hacker Newsnew | past | comments | ask | show | jobs | submit | rsync's commentslogin

Altavista was fantastic and represented a features and usability high water mark that was never passed by google.

Full boolean operator search with "literals" actually respected, negative search terms worked as advertised, etc.

None of that ever worked properly, consistently, at google.


> None of that ever worked properly, consistently, at google.

My experience is it worked pretty well on Google for a while, but then it got progressively worse.


Right, for this first 5 years or so, it worked. But then they started to optimize for “the masses”, and they don’t use boolean logic in queries.

They optimized for ad impressions. There was no technical reason not to keep around a Boolean mode - some competitors effectively exist because of that single feature.

i don't know precisely the architectures they both used (i tend not to study things that are changing and over which i have no control), but here's what I would say:

I like boolean and literalism etc., I like control and syntactic precision, and I did not prefer google when it first got traction and buzz, but within six months of that, google's "page ranked" back-end database was clearly superior to what altavista's front-end queries could do with their own back end data.

it shocked me when people I thought I knew well would say "I always hit google's "I feel lucky" to go straight to the top search result. Me, I prefer to pore through results looking for nuance and to fine tune my query. google was giving me much better results to look at, even if I had less control for fine tuning. Google has relentlessly over time diminished literalism in queries in favor of mass market popularity. As an overly simplistic example, when I look up Thor, I am never interested in any film or who was in it, and that's pretty much all you get now. Alexander the Great is an incredible figure from history, shaping the geo landscape in ways that still affect us today, but searchwise he's just a fictionalized portrayal by a celebrity who don't even have his own authenticity.


You might want to search for "alexander the great" again, and also, maybe use "Alexander IV" or "Alexander of Macedon". I'm an amateur Classicist I look up ancient figures all that time, obscure and well known to check wikipedia on things, and I've never seen it prioritize that film above the figure, though perhaps it did when that movie was recent. Pity about Thor and the MCU, though.

Agreed. AltaVista was the best of the pre-Google search engines. I seem to remember Google having negative terms, literals and booleans (at least or/and) - although they weren't well documented, they worked. Amazon had literals and negative terms too for many years. Now searching on both of those sites is "search theater", where they pretend to give targeted results while burying the result you're looking for just deep enough to maximize page views before too many users bounce.

I fucking hate we now live in a world where leading companies A/B test precisely how much they can degrade their core product value and annoy users knowing they're safe from competitors because startups know if they threaten Google/Amazon on that stuff they'll just put back the minimum functionality long enough to ensure the new player dies.


I pay for kagi on my personal machine, it is always a delight when my cmd-t search is answered kagi and not a list of ads ...

Google did for a long time have a page documenting those features.

Picked a random date from around the time I know they had that. Clicked Adcanced Search, then a link near the top of the page to Advanced Search Tips.

https://web.archive.org/web/20041017053307/http://www.google...


I would think that 90% of the principals at DEC/Compaq WRL working on AltaVista would have moved to google, their first office was nearby in downtown Palo Alto back in 1999.

I used to read through search engine patents back in 2006-7 when I was an SEO consultant. I could see the same names from the AltaVista patents later start appearing as authors on the Google patents.

It is absolutely insane to say that Altavista was better than Google though.

Try Kagi, it implements them quite well.

It worked pretty well on early google and altavista. Find an archive of searchlores.org from that era and see for yourself. +Fravia had documented and tested the features quite thoroughly

The "near" operator was gold.

Altavista was like the 'free' version of what some libraries had for via paid search subscriptions.

At the time where search was a tool that you had to you know.. come up with various terms (remember Google Whacks) and find results about it.

RIP Altavista


I think they also allowed distance between words (within x) to increase relevance.

FWIW, a sharp and immediate uptick in support tickets from self-identified WHM resellers and admins who need backups from their rsync.net accounts right now.

I would think that for everyone that needs some help, there must be 10 who self served…


I don’t think it is.

In fact, all else being equal, switching to work from home should be deflationary.

You spend less on gas, less on eating out, less on movement and activity in general…

The transition to work from home may very well be inflationary… But the end result seems quite obviously deflationary to me.


Wiring ports for humans to use in a flexible and future proof manner (as in a single family home, for instance) gains a lot of utility with PoE.

The convenience and flexibility of PoE would always push me towards copper wiring.


"Well I can't see this ending well. It's either more invasive KYC ..."

I think there is an opportunity here for an elegant solution.

Banks, by definition, know quite a bit about you and aspects of your identity and this is not necessarily problematic nor dangerous.

Further, banks enjoy exorbitant privileges above all other business firms and organizations - privileges that the public rarely receives any upside in exchange.

For these reasons, I think we should consider concentrating KYC responsibilities with the banks such that they do the heavy lifting and the rest of the economy reaps the benefits.

Here is one small example:

A credit or debit card which, by virtue of the card number itself, identifies the user as being over 18 years of age. The bank already knows this information with very high confidence and now smaller, less resourced firms could make use of this to effectively age-gate with almost no investment and no fragmented intrusion into the private lives of their customers.

I don't see any world in which the banks don't have all of this information anyway - why not get some value out of it ?


I would love to live in a world where I don't have to "send a pin to my cell" and instead this task is deferred to my bank.

I’ve said this for as long as I’ve been here on hacker news…

I want the terminal to be as dumb as possible.

I don’t want it to have any understanding of what it is displaying or anscribe any meaning or significance to the character characters it is outputting.

The first time apples terminal.app displayed that little lock icon at the ssh password prompt?

The hairs on the back of your neck should have stood up.


What you’re describing would be a completely unusable terminal. You’d lose things as basic as the backspace key. And what’s wrong with Terminal.app indicating when it’s suppressing output?


Terminal.app does not suppress output in my example.

The ssh command switches the terminal into no-echo mode with termios flags.

Terminal.app, being clever, watches for disabled echo (among other things) and assumes a password is being entered and displays the key icon and enables Secure Event Input.

I don't want Terminal.app to be clever.


Why not?


You're talking nonsense. Backspace worked entirely fine on dumb terminals


"... and that possibly every thought you have ever had and will have is a product of these influences."

Obligatory "Cerulean Top" clip:

https://youtu.be/vL-KQij0I8I


"I guess the problem with Backblaze's business model with respect to Backblaze Personal is that it is "unlimited"."

The new and very interesting problem with their business model is that drive prices have doubled - and in some cases, more than doubled - in the last 12 months.

Backblaze has a lot of debt and at some point the numbers don't make sense anymore.


Yeah, I found that out recently when I had to purchase a new 16TB drive because of them in my RAID died recently. I bought the hard drive used about three years ago for about $130. To replace it I had to shop around and I ended up paying about $270 and I think that was considered a decent deal right now.

Oh well, I guess this is why we're given two kidneys.


I bought a 24TB for £365 last June. It's £905 now.


Is it that bad? When I look at the prices of new drives on amazon I mostly see increases just under 50%. I think used went up more but that's not affecting backblaze as much.


It's pretty bad. The enterprise drives I bought a year ago for under $300 are now either out of stock, or priced above $600 with limited stock.


Thanks for your kind words.

Just to clarify - there are discounted plans that don't have free ZFS snapshots but you can still have them ... they just count towards your quota.

If your files don't change much - you don't have much "churn" - they might not take up any real space anyway.


I don't think a lot of people know you also support Borg 1.x (if you don't absolutely correct me!)

It would be incredible if you started to look into S3 compatible object stores, unless you have made a business decision not to support it.

Thank You for providing an affordable option for self hosters.


Depending on the usecase, rclone can expose an S3 endpoint via `rclone serve s3` to route to another protocol, eg sftp.

I mention it not to shill rsync.net, but to shill rclone, because when I discovered it I was even more impressed with it.

Obviously having to run a command and apply some amount of plumbing is different to a service just providing that API at the outset so the applicability for users will differ but still, rclone is very cool!


We will continue to specialize in filesystem provision, not object storage.

However, we do support interoperating with block storage, such as 's5cmd':

https://news.ycombinator.com/item?id=44248372

... and, of course, rclone, which you can invoke remotely, on our end to move data between cloud accounts, etc.


Thank You for that link!

> not object storage

Happy to email you, if that's better, but is this because of unsustainable competition in the space or the tremendous volatility in consumption that object storage customers bring to the table?

I ask because in this current market, I would imagine investing in storage infrastructure is painful, but then I wonder, you are still in the storage infrastructure space anyways, so it likely has to do with the user behavior or user expectations or both.


Not supporting S3 (and block storage) is not a business decision - it is an ideological decision.

We want to live in a world of UNIX filesystems and we want those to be available in the modern "cloud" ecosystem.


“Only a small minority of users know about settings and how to change them. The vast majority of users do not change default settings.”

Even worse, whatever critical settings you may set as a sophisticated user will frequently be reset, or changed, or re-organized under different settings… And of course, set back to insecure defaults… With subsequent software updates.

This is a regular occurrence with Firefox and privacy settings.

Whatever the actual impetus is, we should act as if this is intentional.


Strictly speaking, any software that cannot easily and permanently disable forced updates is insecure.

So, I don't like auto-updating.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: