WTF Bites
-
@BernieTheBernie said in WTF Bites:
@DogsB What are you complaining about?
That the mug is not dishwater proof? Nothing unusual.
That the mug is not microwave proof? Don't know how common that is; anyways, I do not have a microwave oven.
That no barcode was stuck on the "Barcode Placement" area? So what?
That the thing isMade in China? Why would you expect something different?
It says do not print.
-
@DogsB the all important question, can you expense it?
-
@DogsB the all important question, can you expense it?
Tragically not but you have reminded me to expense something else.
-
Of course the method does have to urlencode the path, because it might contain other characters not valid in a URL. It just shouldn't be encoding the
/s. Shows even such ‘simple’ operation as urlencoding something isn't necessarily simple.The sad thing is, if it specifies it wants just a file name, not a path to a file, it's doing what it should even though that's generally not what you want.
Well, it doesn't, it wants a path to a file relative to the ‘container’. By default, the
/s are not special to the server, but a ‘hierarchical namespace’ can be turned on and then the directories do get some properties. And either way the service is supposed to be able to serve static websites and those usually use paths with/-separated components, so the path is supposed to contain them.I ran into something similar recently with a "website in a box" program with Windows and Android versions (among others). It attempts to write to a file in a folder in a folder relative to a base folder (so "foo/bar/baz.txt"), creating the folders and file if needed.
On Windows you get what was intended: a folder named "foo" containing a folder named "bar" containing a file named "baz.txt".
On Android you get a folder named "foo/", a folder named "foo/bar/", and then either an error if the intended folder structure didn't already exist or a new "baz.txt" if it did.
Android is built on top of Linux, and while that treats directory entry names as bytestrings, and allows everything else,
/is not allowed and not possible to insert, because the kernel always splits the path on/s first.So this sounds like some library somewhere higher up having the
great idea to indicate something is a directory by appending the /to its name, and then the application chokes on that, which it can't be really blamed for, because it's dumb to do that in the first place.
-
:thereifixedit:
-
Google is getting a lot worse because of the Reddit blackouts
/cdn.vox-cdn.com/uploads/chorus_asset/file/24016886/STK093_Google_03.jpg)
A Reddit result in Google might take you to a private page now
This is certainly a personal problem:
With Google’s generally poor search results nowadays, appending “reddit” has long been the default way I search for almost anything
-
@boomzilla my experience with Google is that NOT appending "reddit" to the query yields almost entirely useless results. So with the blackout now, Google is becoming almost entirely useless no matter what I do.
-
@boomzilla my experience with Google is that NOT appending "reddit" to the query yields almost entirely useless results. So with the blackout now, Google is becoming almost entirely useless no matter what I do.
This is certainly a personal problem.
I've never done this unless I was specifically looking for something on reddit (like tracking down a screenshot of a post there) and I generally don't have much of a problem finding what I'm looking for.
What sorts of things are you searching for that is helped by this?
-
@boomzilla said in WTF Bites:
@boomzilla my experience with Google is that NOT appending "reddit" to the query yields almost entirely useless results. So with the blackout now, Google is becoming almost entirely useless no matter what I do.
This is certainly a personal problem.
I've never done this unless I was specifically looking for something on reddit (like tracking down a screenshot of a post there) and I generally don't have much of a problem finding what I'm looking for.
What sorts of things are you searching for that is helped by this?
I’m somewhere in the middle here. I’ve never done that and absolutely wouldn’t expect that to improve search results, but Google results have been steadily going downhill for years.
It used to just give you what you wanted, now it’s mostly bullshit. It can't even prioritize e.g. StackOverflow or Wikipedia above obvious bullshit sites just stealing content from there.E: fixed.
-
@boomzilla said in WTF Bites:
What sorts of things are you searching for that is helped by this?
Reviews, product recommendations, DIY ideas and instructions, workarounds for bugs in popular software, etc. If you don't include "reddit", 99% of results are either sponsored, or content farms with the same copy-paste solutions for every problem.
-
It used to just give you what you wanted, now it’s mostly bullshit. It can even prioritize e.g. StackOverflow or Wikipedia above obvious bullshit sites just stealing content from there.
I suspect you meant "can't"? That's the only way I can make sense of what you wrote...but for me StackOverflow results are pretty much always on the first page if not actually the top. Ditto for wikipedia.
-
@boomzilla it used to be like that for me but some months back, Wikipedia stopped showing up until I specifically search for Wikipedia.
-
@boomzilla said in WTF Bites:
It used to just give you what you wanted, now it’s mostly bullshit. It can even prioritize e.g. StackOverflow or Wikipedia above obvious bullshit sites just stealing content from there.
I suspect you meant "can't"? That's the only way I can make sense of what you wrote...
Yes, of course, typo.
-
content farms
Is that what they're called? There's so many sites that shouldn't even exist. Why doesn't google shove them to where nobody finds them?
- Content blatantly stolen from SO and fucked up.
- "
blenderis a .dll file, which is a kind of system file. Our system rates this files as 65% dangerous". - "Here are 6 different solutions for
show battery indicator, all of which are 95 steps long, written by an Indian, and none of them work". - "Flutter vs React Native differences: Star rating (4.2, 4.8), Values and Ethics: (...), Pricing: free quotation available." (I.e. automatically generated bullshit not even close to relevant)
-
Why doesn't google shove them to where nobody finds them?
Most (if not all) of them get revenue through Google-owned ad services.
-
Why doesn't google shove them to where nobody finds them?
Most (if not all) of them get revenue through Google-owned ad services.
So enshittification has advanced from Facebook to Google search.
-
On Android you get a folder named "foo/", a folder named "foo/bar/", and then either an error if the intended folder structure didn't already exist or a new "baz.txt" if it did.
That was my response. Thankfully I don't have to deal with it; I was just trying to recreate someone else's issue from a forum post to see if I could help them out. For all I know it's not actually forward slashes; they could be a different (Unicode?) slash or the string "%2F" that is "helpfully" being decoded for you.
-
Why doesn't google shove them to where nobody finds them?
Most (if not all) of them get revenue through Google-owned ad services.
So enshittification has advanced from Facebook to Google search.
It's always been in Google search
-
@boomzilla said in WTF Bites:
for me StackOverflow results are pretty much always on the first page if not actually the top. Ditto for wikipedia.
Yeah same here.
On this note - earlier today I searched for something about
gpg. First result was StackOverflow. Second result was StackExchange.
Visiting either showed me a message that StackExchange was down for maintenance and would be back shortly.
Third was a result from AskUbuntu.com and it turns out that's also hosted on StackExchange so was down.
None of the other results even on the first page looked like they answered what I needed so I ignored the problem for a couple of hours and went back when SE was online.
-
my experience with Google is that
NOT appending "reddit" tothe query yields almost entirely useless resultsI've been using duckduckgo¹ for a long time, and left bing as the default in the work browser. Occasionally when neither finds something relevant, I try Google too, and it tends to get even worse results anyway, so there ain't much point.
Wikipedia stopped showing up until I specifically search for Wikipedia.
Both duckduckgo and bing show wikipedia separately with a bigger preview if there is a match, but I also have a wikipedia search engine shortcut and use it whenever I think it's where I want to look first.
¹ Yeah, it's not as private as it touts, but it still gets better results than google and has no account to track me by.
-
Both duckduckgo and bing show wikipedia separately with a bigger preview if there is a match
Yes, I see these in google a lot, too.
-
@loopback0 said in WTF Bites:
Why doesn't google shove them to where nobody finds them?
Most (if not all) of them get revenue through Google-owned ad services.
So enshittification has advanced from Facebook to Google search.
It's always been in Google search
Back in the day, the reason for Google’s success and then dominance was that their search was actually good. Okay, that’s probably 25 years ago by now.
Yes, Google was always about ads, but not to the point of being useless.
-
@topspin there's that and there's the fact that there were fewer content farms back in the day each trying to scratch out revenue.
There was a time when StackOverflow didn't exist (
) and all we could do was fake user-agents to bypass Experts Exchange's no-guest-views policy...
-
Outlook web:
How about you take your
redactedfingers and fuck off.
-
@topspin They're giving you a bunch of tips, and you're complaining?
-
@boomzilla my experience with Google is that NOT appending "reddit" to the query yields almost entirely useless results. So with the blackout now, Google is becoming almost entirely useless no matter what I do.
I have literally never done this and yet the only complaint I have with Google results is that they lie about number of results returned.
-
@TimeBandit said in WTF Bites:
@topspin They're giving you a bunch of tips, and you're complaining?
He wants just one tip.
-
On Android you get a folder named "foo/", a folder named "foo/bar/", and then either an error if the intended folder structure didn't already exist or a new "baz.txt" if it did.
That was my response. Thankfully I don't have to deal with it; I was just trying to recreate someone else's issue from a forum post to see if I could help them out. For all I know it's not actually forward slashes; they could be a different (Unicode?) slash or the string "%2F" that is "helpfully" being decoded for you.
...and now that I've seen it for myself the answer was the obvious one: the slashes were backslashes the whole time.
-
content farms
Is that what they're called? There's so many sites that shouldn't even exist. Why doesn't google shove them to where nobody finds them?
- Content blatantly stolen from SO and fucked up.
- "
blenderis a .dll file, which is a kind of system file. Our system rates this files as 65% dangerous". - "Here are 6 different solutions for
show battery indicator, all of which are 95 steps long, written by an Indian, and none of them work". - "Flutter vs React Native differences: Star rating (4.2, 4.8), Values and Ethics: (...), Pricing: free quotation available." (I.e. automatically generated bullshit not even close to relevant)
They make more money
selling antivirus softwareselling SEO services than they would make from having good search results
-
@boomzilla it used to be like that for me but some months back, Wikipedia stopped showing up until I specifically search for Wikipedia.
Preceding the query with
!wikiworks for me.Wait, different search engine.
-
@Parody Never trust other people to know what backslashes are. Rookie mistake.
-
@Parody Never trust other people
to know what backslashes are. Rookie mistake.
-
Just...everything about this page:
-
@boomzilla Ach. I see what's wrong with it. Needs some parallax scrolling.
-
@Applied-Mediocrity artists gonna art.
-
@boomzilla designed on a big Mac for a Mac (given the resolution) and it's made by people who clearly have gigabit internet connections because this thing takes time to load on my new-house's 360mbps connection.
-
@boomzilla The most helpful part
-
@Parody Never trust other people to know what backslashes are. Rookie mistake.
I can't count the number of times people have argued with me whether a slash was backwards or forwards....
-
@Tsaukpaetra ObXKCD:
-
@Parody Never trust other people to know what backslashes are. Rookie mistake.
Also don't trust the dev to not mess it up, though the fact that it works for the actual file makes me wonder if they have some kind of
leftpad()issue.
-
-
@Tsaukpaetra said in WTF Bites:
For public servers, just install cert-manager (k8s) or dehydrated (linux) or somesuch.
Not a public server. 😉 And if you're exposing the admin interface of your NAS to the public internet, I have a loosely-phallic-shaped object you can interact with.
Status: Couldn't figure out where my old CA stuff was, so re-made it all.
TIL the server does not like signing requests without a special attribute for
CertificateTemplate. Annoying.Anyways, my domain controllers now have certificates!
And the NAS. Not that anything other than the domain computers will want to talk to it anyways.The stupid management interface still doesn't care for that certificate.
What it should be is a nibble simpler. It would be nice if the active directory controller added the ACME protocol, so you could just point any server that is not visible from the outside to your internal ACME endpoint and called it a fortnight like you do with public-facing servers and the public Letsencrypt server.
I was considering making a feature request for this, since it has AD connectivity and does other certificate-things to connect (though probably only incidentally as part of what the samba software does internally) so you'd think it would be possible to automate. Sadly
While standing up the certificate services again, I came across this project, which interfaces with AD CS and might work for some things...
GitHub - glatzert/ACME-Server-ADCS: ACME (RFC 8555) Server compatible implementation, connecting to Active Directory Certificate Services (ADCS)
ACME (RFC 8555) Server compatible implementation, connecting to Active Directory Certificate Services (ADCS) - glatzert/ACME-Server-ADCS
Maybe when I'm bored.
-
@Tsaukpaetra said in WTF Bites:
GitHub - glatzert/ACME-Server-ADCS: ACME (RFC 8555) Server compatible implementation, connecting to Active Directory Certificate Services (ADCS)
ACME (RFC 8555) Server compatible implementation, connecting to Active Directory Certificate Services (ADCS) - glatzert/ACME-Server-ADCS
Unfortunately it says it's not free for commercial use, and the way it's said and the license text in German don't inspire much confidence.
I also noticed there is an ADCS plugin[1][2] for cert-manager[3] (which I usually use anyway), but it is unfortunately no longer maintained.
Well, if you have one server, getting a new cert every other year from the web UI is acceptable. And when I needed a bunch of them for a project, I got a subordinate CA with name restriction signed by that UI and gave that to the cert-manager to sign all the project subdomains—because in DNS and in name restriction the
*is recursive, but in certificates it isn't, so one *.project.domain certificate does not always cut it.
-
Unfortunately it says it's not free for commercial use
Unless your company has a total turnover of less than 10 million €.
Probably not relevant to you, just wanted to clarify.1500€ doesn't strike me as much, but then I have no idea if this thing is worth a single penny.
-
@topspin 1500 is in the ballpark of 3 man-days (we are cheap here in Czechia) and it will take a lot of servers before that thing saves 3 man-days for the admins, though it might raise if the certificates expiring ends up blocking a lot of people.
-
*grabs laptop that I haven't used in a while*
Might be good to run through Windows Updates to avoid problems later down the road, when there isn't a stable interweb connection.
- Run Windows Updates manually
Checking for updates...
(10 minutes later)
Checking for updates...
(20 minutes later)
Checking for updates...
(40 minutes later)
Checking for updates...
- Restart machine
Restarting...
(5 minutes later)
Restarting...
(10 minutes later)
Restarting...
Fucking Windows.
Edit: It restarted now (I was googling how to reset it - there aren't any physical buttons). After booting up, it's now downloading an update ever soooooooo slowly. And it showed me a toaster about Microsoft Rewards.
- Run Windows Updates manually
-
This post is deleted!
-
-
@HardwareGeek It's a bit like dealing with any hazardous material. If possible, you want to do it in a controlled environment and before it becomes an emergency.
-
@HardwareGeek It's a bit like dealing with any hazardous material. If possible, you want somebody else, far away, to do it in a controlled environment and before it becomes an emergency.
