Is that even possible o_O



  •  I'm currently seeing 1 password protected and 15 unsecured wi-fi networks being broadcast in this apartment building alone (which houses mostly old people...). And then there's a mysterious network that's named after a university building 1.4KM away.

     Granted I do have line of sight of that building, but is that even possible?

    Of course someone could have bought or stolen some hardware from that place. That seems almost equally unlikely.

    What to think of this?



  • @Brother Laz said:

    And then there's a mysterious network that's named after a university building 1.4KM away.

    Without directional antennas it's not coming from the building; but if you are close to a university and in an apartment building then it could be the name someone picked for their network when configuring it (lots of students pick things with their school when doing network names, I live in a complex of mostly students and grad students where lots of network names come from stuff at the uni).



  • Are they all proper networks or could they be adhoc? There was a bug in windows xp that if it couldn't connect to an ap it would create an adhoc network with the same name. Google "Free Public WiFi".



  • Nothing special, many universities experimented with wide area wifi networks and have directional antennas on the roof of buildings. Probably you're living in between two university buildings.


  • Trolleybus Mechanic

    @Zemm said:

    Are they all proper networks or could they be adhoc? There was a bug in windows xp that if it couldn't connect to an ap it would create an adhoc network with the same name. Google "Free Public WiFi".
     

    Fuck, is that what Free Public WiFi was?  I always assumed it was some rouge AP or a laptop infected with malware trying to lure other victims.


  • Discourse touched me in a no-no place

    @Lorne Kates said:

    I always assumed it was some rouge AP or a laptop infected with malware trying to lure other victims.
    Trying to lure you into using Windows? Probably…



  • @Lorne Kates said:

    @Zemm said:

    Are they all proper networks or could they be adhoc? There was a bug in windows xp that if it couldn't connect to an ap it would create an adhoc network with the same name. Google "Free Public WiFi".
     

    Fuck, is that what Free Public WiFi was?  I always assumed it was some <font color="#FF0000">rouge AP</font> or a laptop infected with malware trying to lure other victims.

    Why would you assume the AP was red?

     



  • And then there's a mysterious network that's named after a university building 1.4KM away.

    Is there a network of the same name in the actual building? It's possible that there's a clever geek with Wireshark running in your apartment, hoping that someday somebody with his mobile configured to auto-connect will walk past your building while checking his unencrypted bank account.



  • @Maciejasjmj said:

    And then there's a mysterious network that's named after a university building 1.4KM away.

    Is there a network of the same name in the actual building?

    Why not? You can call it anything you want. I never gave it much thought and always used the router's default name, until I moved into a new neighborhood and there were 6 or 7 connections all called "Linksys". Now mine is called "FBI Surveillance".

     



  • @El_Heffe said:

    @Maciejasjmj said:

    And then there's a mysterious network that's named after a university building 1.4KM away.

    Is there a network of the same name in the actual building?

    Why not? You can call it anything you want. I never gave it much thought and always used the router's default name, until I moved into a new neighborhood and there were 6 or 7 connections all called "Linksys". Now mine is called "FBI Surveillance".

     

    Isn't the "autoconnect to known networks" feature based on SSIDs? If you have LINKSYS remembered, and travel to your friend who also has a network named LINKSYS, won't the phone/laptop try to connect? Just asking, never really used that feature.



  • @El_Heffe said:

    [quote user="Lorne Kates"]

    [quote user="Zemm"]Are they all proper networks or could they be adhoc? There was a bug in windows xp that if it couldn't connect to an ap it would create an adhoc network with the same name. Google "Free Public WiFi".

     

    Fuck, is that what Free Public WiFi was?  I always assumed it was some <font color="#FF0000">rouge AP</font> or a laptop infected with malware trying to lure other victims.

    [/quote] Why would you assume the AP was red?

     

    [/quote]
    Rouge

    Rougelike



  • @Ben L. said:

    @El_Heffe said:

    @Lorne Kates said:

    @Zemm said:

    Are they all proper networks or could they be adhoc? There was a bug in windows xp that if it couldn't connect to an ap it would create an adhoc network with the same name. Google "Free Public WiFi".
     

    Fuck, is that what Free Public WiFi was?  I always assumed it was some <font color="#FF0000">rouge AP</font> or a laptop infected with malware trying to lure other victims.

    Why would you assume the AP was red?

     

    Rouge
    Rougelike
    <font color="#fec2ca" size="7">Rouge Lite</font>

     



  • @El_Heffe said:

    @Ben L. said:

    @El_Heffe said:

    @Lorne Kates said:

    @Zemm said:

    Are they all proper networks or could they be adhoc? There was a bug in windows xp that if it couldn't connect to an ap it would create an adhoc network with the same name. Google "Free Public WiFi".
     

    Fuck, is that what Free Public WiFi was?  I always assumed it was some <font color="#FF0000">rouge AP</font> or a laptop infected with malware trying to lure other victims.

    Why would you assume the AP was red?

     


    Rouge

    Rougelike
    <font color="#fec2ca" size="7">Rouge Lite</font>



  • @Brother Laz said:

    Granted I do have line of sight of that building, but is that even possible?
    Sure.



  • @Brother Laz said:

    building 1.4KM away.

     Granted I do have line of sight of that building, but is that even possible?

    I once had a solid 802.11b link over 4km long. It was great: transfers were around 600 KBytes/s and no lag: much faster than any broadband available to us at the time. We played a lot of Battlefield 1942. But we did have directional antennae on both ends.



  • @Brother Laz said:

     I'm currently seeing 1 password protected and 15 unsecured wi-fi networks being broadcast in this apartment building alone (which houses mostly old people...). And then there's a mysterious network that's named after a university building 1.4KM away.

     Granted I do have line of sight of that building, but is that even possible?

    Of course someone could have bought or stolen some hardware from that place. That seems almost equally unlikely.

    What to think of this?

    It could also be someone who uses the same SSID to sniff passwords and such. It is actually very simple: make a network with the name eduroam (if you are in a place with a lot of students) or "t-mobile" (or your regional equivalent, in the Netherlands they offer free wifi in trains) with no auth and sniff the network. Voila, you now have logins of services that don't use encryption(passwords for internet banking and email will most likely be the same). Also, whatsapp and snapchat messages and the like.

    Someone I know once anonymously sent a list of ~100 student logins (+ a few faculty staff) to show to the uni ict staff that it was a very stupid idea to have their default and recommended eduroam settings to not verify the RADIUS server authenticity (certificate) in combination with most of the internal web services not using SSL or some other form of encryption. He just plugged his laptop into the uni network in the library, made a virtual access point and captured and analysed all traffic.



  • @Maciejasjmj said:

    Isn't the "autoconnect to known networks" feature based on SSIDs? If you have LINKSYS remembered, and travel to your friend who also has a network named LINKSYS, won't the phone/laptop try to connect?
    Yes and yes. It's kind of handy because the iPhone's system of managing wireless networks is awful: you have to be within range of a network in order to make your phone "forget" it. Otherwise, it'll remember the network forever. But since it'll try to join any wireless AP based on the SSID, you can change your AP's SSID to the one you're trying to make your phone forget about, tell it to forget it, and then change the SSID back.



  • @Maciejasjmj said:

    Isn't the "autoconnect to known networks" feature based on SSIDs? If you have LINKSYS remembered, and travel to your friend who also has a network named LINKSYS, won't the phone/laptop try to connect? Just asking, never really used that feature.
     

    And then you go to another friend's house, where he uses a password different from your own, and when you come back you'll have to configure it again, every time.

    It's just easier to use a more unique identifier.

     


  • Discourse touched me in a no-no place

    @Mcoder said:

    And then you go to another friend's house, where he uses a password different from your own, and when you come back you'll have to configure it again, every time.

    It's just easier to use a more unique identifier.

    I took to naming mine (rather unoriginally) Police Surveillance Van #42. Until I got sick of (I presume - couldn't 100% confirm it) some wiseacre trying to mess around with it.



    Now it's just the default virginmedia[numbers] (and additional virginmedia[same numbers]_guest for when I do have friends round.) Idiot seems to have stopped playing around with it. I've had less problems with it anyway.



  • @PJH said:

    I took to naming mine (rather unoriginally) Police Surveillance Van #42. Until I got sick of (I presume - couldn't 100% confirm it) some wiseacre trying to mess around with it.
    Can SSIDs be Unicode?  Maybe I should change my network to something in Arabic.



  • @da Doctah said:

    @PJH said:

    I took to naming mine (rather unoriginally) Police Surveillance Van #42. Until I got sick of (I presume - couldn't 100% confirm it) some wiseacre trying to mess around with it.
    Can SSIDs be Unicode?  Maybe I should change my network to something in Arabic.

    ␇م␇ا␇ ␇ا␇ل␇ل␇ع␇نة␇



  • @da Doctah said:

    Can SSIDs be Unicode?  Maybe I should change my network to something in Arabic.
    Yes and no. The SSID doesn't have any specific encoding type, so you might think that the default encoding would be ASCII, which does not support Unicode (and indeed some clients do treat it as ASCII). However, some clients treat it as UTF-8, which allows it to have Unicode (and even Emoji, which is just a subset of Unicode really). A SSID with Unicode characters will only appear correctly in a client that treats the SSID as UTF-8 encoded text. Clients which treat it as ASCII text will show garbage characters where the Unicode characters should be.



  • @anotherusername said:

    The SSID doesn't have any specific encoding type, so you might think that the default encoding would be ASCII

    Why would you think that? Nothing made after 1995 should use ASCII as a default.



  • @anotherusername said:

    It's kind of handy because the iPhone's system of managing wireless networks is awful: you have to be within range of a network in order to make your phone "forget" it. Otherwise, it'll remember the network forever.

    Is that true? I've never had an iPhone but that seems... oh, how to put this... really fucking stupid.



  • @anonymous234 said:

    @anotherusername said:
    The SSID doesn't have any specific encoding type, so you might think that the default encoding would be ASCII

    Why would you think that? Nothing made after 1995 should use ASCII as a default.

    shrug


    @2.2.1.1.7 802.11 SSID said:
    [b]SSID_String (variable):[/b] The ASCII representation of the SSID for the basic service set with which a wireless responder's wireless network interface associates.

    At some point, someone at Microsoft decided that the SSID should be ASCII. Some other systems treat it as UTF-8.



  • @mikeTheLiar said:

    @anotherusername said:
    It's kind of handy because the iPhone's system of managing wireless networks is awful: you have to be within range of a network in order to make your phone "forget" it. Otherwise, it'll remember the network forever.

    Is that true? I've never had an iPhone but that seems... oh, how to put this... really fucking stupid.

    Yes. (unless you jailbreak)

    There is no way to manage the list of known networks. If you're within range of a particular network, it'll appear in the list of available networks, and you can view its properties and select "Forget this Network".


  • Considered Harmful

    @Ben L. said:

    @da Doctah said:

    @PJH said:

    I took to naming mine (rather unoriginally) Police Surveillance Van #42. Until I got sick of (I presume - couldn't 100% confirm it) some wiseacre trying to mess around with it.
    Can SSIDs be Unicode?  Maybe I should change my network to something in Arabic.

    ␇م␇ا␇ ␇ا␇ل␇ل␇ع␇نة␇

    سمَـَّوُوُحخ ̷̴̐خ ̷̴̐خ ̷̴̐خ امارتيخ ̷̴̐خ



  • @mikeTheLiar said:

    Is that true? I've never had an iPhone but that seems... oh, how to put this... really fucking stupid.
    Windows 8 has the same "feature".



  • @ender said:

    @mikeTheLiar said:
    Is that true? I've never had an iPhone but that seems... oh, how to put this... really fucking stupid.
    Windows 8 has the same "feature".

    But unlike Apple devices, Windows 8 does have built-in wireless profile management from the command-line and allows you to use third-party wireless profile management utilities without voiding your warranty, so if you're geeky enough to actually care about your device's remembered wireless profiles, you're not completely SOL.



  • @anonymous234 said:

    @anotherusername said:
    The SSID doesn't have any specific encoding type, so you might think that the default encoding would be ASCII

    Why would you think that? Nothing made after 1995 should use ASCII as a default.

    I don't think anyone should use Unicode without specific reason to do so, e.g. some sort of specific multi-language requirement. It's a mess. I've already thoroughly deconstructed it in another post on this site. The adoption of the UTF-8 encoding basically represented Eunuch-code proponents surrendering to the 90%+ of us who just want to use ASCII.


  • Discourse touched me in a no-no place

    @bridget99 said:

    I don't think anyone should use Unicode without specific reason to do so, e.g. some sort of specific multi-language requirement. It's a mess. I've already thoroughly deconstructed it in another post on this site. The adoption of the UTF-8 encoding basically represented Eunuch-code proponents surrendering to the 90%+ of us who just want to use ASCII.
    The alternatives mostly suck worse. UTF-8 has the advantage of being ASCII-friendly and yet letting you pass around moonrunes if you need them. Outside the English-speaking bits of North America (and the parts of Australia and New Zealand that aren't sheep and kangaroos, but who cares about them?) non-ASCII is very useful. UTF-8 is so much better than the crap with code page diddling that we used to put up with.



  • I need moonrunes.



  • @blakeyrat said:

    I need moonrunes.

    RSTFY



  • @anotherusername said:

    @anonymous234 said:
    @anotherusername said:
    The SSID doesn't have any specific encoding type, so you might think that the default encoding would be ASCII
    Why would you think that? Nothing made after 1995 should use ASCII as a default.

    *shrug*

    [quote user="2.2.1.1.7 802.11 SSID"]SSID_String (variable): The ASCII representation of the SSID for the basic service set with which a wireless responder's wireless network interface associates.

    At some point, someone at Microsoft decided that the SSID should be ASCII. Some other systems treat it as UTF-8.

    [/quote] I looked this up when Doctor Who did the wi-fi thing episode.  The original release of the 802.11 spec just said that SSIDs were 32 arbitrary octets and made no mention of encoding.  There was an updated release of the spec a few years later where they amended it to specify that it should be interpreted as UTF-8.




  • @dkf said:

    @bridget99 said:
    I don't think anyone should use Unicode without specific reason to do so, e.g. some sort of specific multi-language requirement. It's a mess. I've already thoroughly deconstructed it in another post on this site. The adoption of the UTF-8 encoding basically represented Eunuch-code proponents surrendering to the 90%+ of us who just want to use ASCII.
    The alternatives mostly suck worse. UTF-8 has the advantage of being ASCII-friendly and yet letting you pass around moonrunes if you need them. Outside the English-speaking bits of North America (and the parts of Australia and New Zealand that aren't sheep and kangaroos, but who cares about them?) non-ASCII is very useful. UTF-8 is so much better than the crap with code page diddling that we used to put up with.

    Yeah, that codepage crap was bad... what I'm suggesting, though, is that internationalization is a quagmire that developers should avoid unless there's a very compelling reason. And "users of our company web site who are in Albania will feel more comfortable if 'Log In', 'Home', etc. are rendered in Albanian" is not a very compelling reason, in my experience. You should use Unicode if you're a philologist working with Albanian text, perhaps, but end users in Albania are fine with seeing (and would even prefer to see) the plain old English field labels we all see. The only real exception I can think of is that some government-related work requires certain languages to be supported. If you're working in that sort of millieu, though, then my attempt to invoke reason here is probably unwelcome. You've already crossed through a magic mirror that I refuse to step through myself.



    I've seen this over-internationalization antipattern in practice several times. A young, enthusiastic web developer goes to heroic lengths to insert internationalization into some otherwise-trivial task, and it turns into a mess / wastes resources / annoys the shit out of people. Many other languages just don't have widely agreed upon terminology for computer-related objects and tasks. And even for those that do, do you really want to expend resources figuring out this terminology, for all the languages you would have to support? After all, if you support Serbocroatian but not Albanian, you may end up being perceived as some sort of Albanophobe zealot. Better to just stick to English and avoid all sorts of controversy (as people have been finding out for all sorts of things since well before people started using computers).



    Many problem domains (e.g. aviation, marine navigation, etc.) are purely English-driven in practice. You don't want to internationalize the software that runs a boat or airplane. English is also the language of the Internet... if you've ever stumbled onto a single post in a comments thread that was stubbornly written in some other language and thought "what a dumbass; no one here is going to understand that" then you know what I mean. This is not xenophobia on my part; frankly, I think English grammar is a mess, English phonology sounds bad, and English orthography is utter stupidity. But it doesn't matter. English is the best and only language appropriate for >90% of software development.



    There are also some functional problems with Unicode that, I believe, tend to relegate it to exactly the sort of niche applications I suggest. You don't want to use it for URLs, for example, because there are myriad characters that look the same but are not, e.g. Angstrom sign vs. "A"-with-circle-above-it. There's no damned reason those should be distinct "code points", but they are, and the result is that Unicode-based URLs are very easy to spoof.


  • Discourse touched me in a no-no place

    @bridget99 said:

    There are also some functional problems with Unicode that, I believe, tend to relegate it to exactly the sort of niche applications I suggest. You don't want to use it for URLs, for example, because there are myriad characters that look the same but are not, e.g. Angstrom sign vs. "A"-with-circle-above-it. There's no damned reason those should be distinct "code points", but they are, and the result is that Unicode-based URLs are very easy to spoof.
    The Unicode Consortium have been fighting back and forth over this for ages. OTOH, problems about when dealing with letters that look very similar but which are fundamentally different. For example, “p” and “р”; the first is a latin lower-case letter and the second is a cyrillic lower-case letter for a completely different consonant, with a different meaning and different sort order. What ever you do, it's not going to be right for some important cases.

    The only big thing in favour of Unicode (and especially UTF-8) is that it is much less bad than the awful mess that preceded it. What an endorsement…


  • Garbage Person

    @bridget99 said:

    I've seen this over-internationalization antipattern in practice several times. A young, enthusiastic web developer goes to heroic lengths to insert internationalization into some otherwise-trivial task, and it turns into a mess / wastes resources / annoys the shit out of people. Many other languages just don't have widely agreed upon terminology for computer-related objects and tasks. And even for those that do, do you really want to expend resources figuring out this terminology, for all the languages you would have to support? After all, if you support Serbocroatian but not Albanian, you may end up being perceived as some sort of Albanophobe zealot. Better to just stick to English and avoid all sorts of controversy (as people have been finding out for all sorts of things since well before people started using computers).
    I had to stop the FNG from internationalizing an internal tool. An internal tool with an audience of 50 people in Pennsylvania, Illinois and Minnesota.

    The Minnesotans have a shitty accent, but they don't speak another damned language.

     

    Bonus: The change he was working on was "Add some validation to that form field to prevent them from typing a local path - that'll write to the server's local disks. Allow only UNC paths." Which you'll note has nothing in any way to do with "fundamentally change the way every string in the application is drawn"



  • How many ways can I say the number one (not counting language-specific characters)?

    U+31
    U+B9
    U+2081
    U+215F
    U+2160
    U+2170
    U+2460
    U+2474
    U+2488
    U+24F5
    U+2776
    U+2780
    U+278A
    U+FF11
    U+1D7CF
    U+1D7D9
    U+1D7E3
    U+1D7E3
    U+1D7F7
    U+E0031

    At this point, I stopped counting.



  • @dkf said:

    OTOH, problems about when dealing with letters that look very similar but which are fundamentally different.

    Or even characters that aren't that different. Just today I found a utf8 bug where the client had an "en-dash" in their company name which caused a Software Error. (Poorly written library that brings a few PHP functions into perl. Luckily it was written in perl (not XS or something) and I found an undocumented function that could handle utf8)



  • @Weng said:

    I had to stop the FNG from internationalizing an internal tool. An internal tool with an audience of 50 people in Pennsylvania, Illinois and Minnesota.

    The Minnesotans have a shitty accent, but they don't speak another damned language.

     

    Bonus: The change he was working on was "Add some validation to that form field to prevent them from typing a local path - that'll write to the server's local disks. Allow only UNC paths." Which you'll note has nothing in any way to do with "fundamentally change the way every string in the application is drawn"



    While I can't comment on the exact situation at your workplace, Minnesota actually has the highest population of Somalis in North America. Not to mention the possiblity of hispanic immigrants, Pennsylvania Dutch, e.t.c.

    Personally, I'm with the FNG on that one. Not that he should have gone rogue and started working on a different task than the one he was assigned, but I agree with him that fixing a string encoding method that is practically guaranteed to cause problems in the future is a priority. I had to deal with intermittent issues for months with people copy-pasting text until I finally just had every string I could touch go through a forced conversion. The problem isn't that someone plans to type arabic into the field. It's that they got a document from someone who worked in a system that allows arabic, copypasted a section of the document and voila, conversion problem. Or someone decides to be super accurate and put the umlaut in HaagenDas or something.

     



  • @Snooder said:

    @Weng said:
    I had to stop the FNG from internationalizing an internal tool. An internal tool with an audience of 50 people in Pennsylvania, Illinois and Minnesota.

    The Minnesotans have a shitty accent, but they don't speak another damned language.

     

    Bonus: The change he was working on was "Add some validation to that form field to prevent them from typing a local path - that'll write to the server's local disks. Allow only UNC paths." Which you'll note has nothing in any way to do with "fundamentally change the way every string in the application is drawn"



    While I can't comment on the exact situation at your workplace, Minnesota actually has the highest population of Somalis in North America. Not to mention the possiblity of hispanic immigrants, Pennsylvania Dutch, e.t.c.

    If they are working for a US company with offices only in the US, as Weng's post seems to indicate, their employer very likely expects them to use English, wherever they may have immigrated from.

    BTW, Pennsylvania Dutch are not Dutch. They are descencents of German (Deutch) immigrants in the 17th and 18th centuries; they've been here for over 200 years. Only Old Order Amish and Old Order Mennonites (who are very unlikely to have IT jobs) use the Pennsylvania Dutch language routinely, and virtually all of them are bilingual in English; many younger Pennsylvania Dutch do not speak it at all.
    @Snooder said:

    Personally, I'm with the FNG on that one. Not that he should have gone rogue and started working on a different task than the one he was assigned, but I agree with him that fixing a string encoding method that is practically guaranteed to cause problems in the future is a priority. I had to deal with intermittent issues for months with people copy-pasting text until I finally just had every string I could touch go through a forced conversion. The problem isn't that someone plans to type arabic into the field. It's that they got a document from someone who worked in a system that allows arabic, copypasted a section of the document and voila, conversion problem. Or someone decides to be super accurate and put the umlaut in HaagenDas or something.
    I agree that a modern application should be able to handle someone pasting in a non-ASCII character (if only we could agree on an encoding), but the context of Weng's post was a discussion of translating strings into other languages. I could be wrong, but I gathered that the FNG was adding this, not merely encoding/decoding.

     


Log in to reply