Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #50524 > unrolled thread

GeoIP2 for retrieving city and region ?

Back to article view | Back to comp.lang.python

Contents

  GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-12 17:18 +0300
    Re: GeoIP2 for retrieving city and region ? Dave Angel <davea@davea.name> - 2013-07-12 11:32 -0400
      Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-12 18:52 +0300
        Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-12 19:32 +0300
          Re: GeoIP2 for retrieving city and region ? Grant Edwards <invalid@invalid.invalid> - 2013-07-12 16:38 +0000
            Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-13 02:47 +1000
            Re: GeoIP2 for retrieving city and region ? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2013-07-12 19:04 -0400
              Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 08:48 +0300
                Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-13 15:53 +1000
                  Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 09:07 +0300
                    Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-13 16:22 +1000
                      Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 09:28 +0300
                    Re: GeoIP2 for retrieving city and region ? Lele Gaifax <lele@metapensiero.it> - 2013-07-13 10:12 +0200
                    Re: GeoIP2 for retrieving city and region ? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2013-07-13 12:48 -0400
                    Re: GeoIP2 for retrieving city and region ? Roy Smith <roy@panix.com> - 2013-07-13 12:59 -0400
                    Re: GeoIP2 for retrieving city and region ? Wayne Werner <wayne@waynewerner.com> - 2013-07-13 13:35 -0500
            Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-13 10:36 +1000
            Re: GeoIP2 for retrieving city and region ? Joel Goldstick <joel.goldstick@gmail.com> - 2013-07-12 19:42 -0400
          Re: GeoIP2 for retrieving city and region ? Ian Kelly <ian.g.kelly@gmail.com> - 2013-07-12 11:19 -0600
          Re: GeoIP2 for retrieving city and region ? Terry Reedy <tjreedy@udel.edu> - 2013-07-12 18:07 -0400
          Re: GeoIP2 for retrieving city and region ? MRAB <python@mrabarnett.plus.com> - 2013-07-12 23:07 +0100
            Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 08:32 +0300
              Re: GeoIP2 for retrieving city and region ? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2013-07-13 12:54 -0400
                Re: GeoIP2 for retrieving city and region ? Roy Smith <roy@panix.com> - 2013-07-13 13:09 -0400
                Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 20:43 +0300
                  Re: GeoIP2 for retrieving city and region ? Roy Smith <roy@panix.com> - 2013-07-13 14:19 -0400
                    Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 21:23 +0300
                      Re: GeoIP2 for retrieving city and region ? Roy Smith <roy@panix.com> - 2013-07-13 14:30 -0400
                      Re: GeoIP2 for retrieving city and region ? Michael Torrie <torriem@gmail.com> - 2013-07-13 16:57 -0600
                        Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-16 22:43 +0300
                          Re: GeoIP2 for retrieving city and region ? Joel Goldstick <joel.goldstick@gmail.com> - 2013-07-16 15:51 -0400
                          Re: GeoIP2 for retrieving city and region ? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2013-07-16 20:51 -0400
                          Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-17 10:58 +1000
                      Re: GeoIP2 for retrieving city and region ? Tim Chase <python.list@tim.thechases.com> - 2013-07-14 13:32 -0500
                  Re: GeoIP2 for retrieving city and region ? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2013-07-13 14:21 -0400
                    Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-13 21:28 +0300
                      Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-14 09:12 +1000
                  Re: GeoIP2 for retrieving city and region ? Wayne Werner <wayne@waynewerner.com> - 2013-07-13 13:46 -0500
                  Re: GeoIP2 for retrieving city and region ? Benjamin Kaplan <benjamin.kaplan@case.edu> - 2013-07-13 11:17 -0700
                    Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-15 16:25 +0300
                      Re: GeoIP2 for retrieving city and region ? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2013-07-15 20:34 -0400
                      Re: GeoIP2 for retrieving city and region ? Michael Torrie <torriem@gmail.com> - 2013-07-15 20:48 -0600
                        Re: GeoIP2 for retrieving city and region ? alex23 <wuwei23@gmail.com> - 2013-07-17 10:07 +1000
              Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-14 03:03 +1000
          Re: GeoIP2 for retrieving city and region ? Michael Torrie <torriem@gmail.com> - 2013-07-12 16:15 -0600
        Re: GeoIP2 for retrieving city and region ? Joel Goldstick <joel.goldstick@gmail.com> - 2013-07-12 12:24 -0400
      Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-14 08:18 +0300
        Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-14 15:24 +1000
          Re: GeoIP2 for retrieving city and region ? Νικόλας <nikos@superhost.gr> - 2013-07-14 10:51 +0300
    Re: GeoIP2 for retrieving city and region ? Chris Angelico <rosuav@gmail.com> - 2013-07-13 02:41 +1000

Page 1 of 3  [1] 2 3  Next page →

#50524 — GeoIP2 for retrieving city and region ?

Hello, iam still looking for a way to identify the city of my website 
visitors.

The closet i have gone is to come up with the visitor's ISP city:

try:
	gi = pygeoip.GeoIP('/usr/local/share/GeoLiteCity.dat')
	city = gi.time_zone_by_addr( os.environ['HTTP_CF_CONNECTING_IP'] )
	host = socket.gethostbyaddr( os.environ['HTTP_CF_CONNECTING_IP'] )[0]
except Exception as e:
	host = repr(e)


But today i was searching again for this and found out about geoip2, 
maybe that would help more.


 >>> import geoip2
Traceback (most recent call last):
   File "<stdin>", line 1, in <module>
ImportError: No module named 'geoip2'


     >>> client = geoip2.webservices.Client(42, 'abcdef123456')
     >>> omni = client.omni('24.24.24.24')
     >>> country = omni.country
     >>> print(country.iso_code)


I cant even import the module even though my 'pip install geopip2' was 
successful

There is definately i way to identify the users location based solely on 
its ip address as this site does it: http://www.geoiptool.com/

Google, MS, facebook and twitter are not the only ones that can do it?

Perhaps this is being done by giving  longitude and latitude?
-- 
What is now proved was at first only imagined!

[toc] | [next] | [standalone]

#50531

On 07/12/2013 10:18 AM, Νικόλας wrote:
> Hello, iam still looking for a way to identify the city of my website
> visitors.
>
    <SNIP>
>
>
> I cant even import the module even though my 'pip install geopip2' wa
> successful

Either it wasn't successful, or it's not the package you thought.  There 
are lots of things you might have downloaded, but since you give no 
details...


>
> There is definately i way to identify the users location based solely on
> its ip address as this site does it: http://www.geoiptool.com/
>

Sure, and as long as you don't mind it being 1000 miles off, you too can 
claim to do it too.  When I go to that site, the little pin is in 
Kansas, which is 1100 miles from where I live on the east coast of the US.


> Google, MS, facebook and twitter are not the only ones that can do it?
>
> Perhaps this is being done by giving  longitude and latitude?

Or by reading the mind of the programmer.

I suggest you read that geoiptool site, in particular the page

http://www.geoiptool.com/en/ip_info/

There is some misinformation, but notice carefully the part about 
dynamic IP addresses.  Probably 99% of the individual users on the web 
(the ones using a browser) have dynamic IP addresses.  The fixed ones 
are needed by servers, and especially for DNS use, where the name lookup 
wants to be stable for relatively log periods of time.

-- 
DaveA

[toc] | [prev] | [next] | [standalone]

#50536

Στις 12/7/2013 6:32 μμ, ο/η Dave Angel έγραψε:
>
> I suggest you read that geoiptool site, in particular the page
>
> http://www.geoiptool.com/en/ip_info/
>
> There is some misinformation, but notice carefully the part about
> dynamic IP addresses.  Probably 99% of the individual users on the web
> (the ones using a browser) have dynamic IP addresses.  The fixed ones
> are needed by servers, and especially for DNS use, where the name lookup
> wants to be stable for relatively log periods of time.


I did, for me it gives exact city location and not the ISP's city location.

I dont know whay for you ti just says Kansas, it shoudln't, since it 
susing longitute and latitude, it should have been accurate.

-- 
What is now proved was at first only imagined!

[toc] | [prev] | [next] | [standalone]

#50541

I know i have asked before but hwta i get is ISP city not visitors 
precise city.

GeoLiteCity.dat isnt accurate that's why it comes for free.
i must somehow get access to GeoIPCity.dat which is the full version.

And of course it can be done, i dont want to believe that it cant.

When visiting http://www.geoiptool.com/en/__ip_info/ it pinpoints my 
_exact_ city of living, not the ISP's.
It did not even ask me to allow a geop ip javascript to run it present 
sit instantly.

So, it certainly is possible if only one can find the correct database 
to use.

So, my question now is, if there is some way we can get an accurate Geo 
City database.

-- 
What is now proved was at first only imagined!

[toc] | [prev] | [next] | [standalone]

#50542

On 2013-07-12, ?????????????? <nikos@superhost.gr> wrote:

> I know i have asked before but hwta i get is ISP city not visitors 
> precise city.

You can't reliably do that.

> GeoLiteCity.dat isnt accurate that's why it comes for free. i must
> somehow get access to GeoIPCity.dat which is the full version.
>
> And of course it can be done, i dont want to believe that it cant.

Believe what you want.

> When visiting http://www.geoiptool.com/en/__ip_info/ it pinpoints my 
> _exact_ city of living, not the ISP's. It did not even ask me to
> allow a geop ip javascript to run it present sit instantly.

So you've reached your conclusion on a sample size of one?

-- 
Grant Edwards               grant.b.edwards        Yow! I'm encased in the
                                  at               lining of a pure pork
                              gmail.com            sausage!!

[toc] | [prev] | [next] | [standalone]

#50544

On Sat, Jul 13, 2013 at 2:38 AM, Grant Edwards <invalid@invalid.invalid> wrote:
> On 2013-07-12, ?????????????? <nikos@superhost.gr> wrote:
>> When visiting http://www.geoiptool.com/en/__ip_info/ it pinpoints my
>> _exact_ city of living, not the ISP's. It did not even ask me to
>> allow a geop ip javascript to run it present sit instantly.
>
> So you've reached your conclusion on a sample size of one?

This is Nikos. He doesn't read responses properly, doesn't do his
research, and has (by his own admission) an iron head that doesn't let
information cross it lightly. Yes, he reached his conclusion on a
sample size of one.

Oh, and just for laughs, I tried a few of my recent mobile IP
addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
some in the CBD and some out in the suburbs, but all vastly wrong, and
places I haven't been. But I'd never expect it to be accurate on
those.

ChrisA

[toc] | [prev] | [next] | [standalone]

#50562

On Sat, 13 Jul 2013 02:47:38 +1000, Chris Angelico <rosuav@gmail.com>
declaimed the following:

>
>Oh, and just for laughs, I tried a few of my recent mobile IP
>addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
>some in the CBD and some out in the suburbs, but all vastly wrong, and
>places I haven't been. But I'd never expect it to be accurate on
>those.
>
	Well... the MaxMind demo of "my IP" did get the proper metropolitan
area... But they list the ISP as "AT&T"... My real ISP is Earthlink
(piggybacking on AT&T DSL service).

	The Lat/Long, however shows as 

42.9634 -85.6681
whereas a recent GPS readout shows
42.9159 -85.5541

or 2m50s too far north, and 6m50s too far west.

	Same website, accessed from my Blackberry phone, gave a result of
"United States, NA" and location 38 -97
-- 
	Wulfraed                 Dennis Lee Bieber         AF6VN
    wlfraed@ix.netcom.com    HTTP://wlfraed.home.netcom.com/

[toc] | [prev] | [next] | [standalone]

#50572

Στις 13/7/2013 2:04 πμ, ο/η Dennis Lee Bieber έγραψε:
> On Sat, 13 Jul 2013 02:47:38 +1000, Chris Angelico <rosuav@gmail.com>
> declaimed the following:
>
>>
>> Oh, and just for laughs, I tried a few of my recent mobile IP
>> addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
>> some in the CBD and some out in the suburbs, but all vastly wrong, and
>> places I haven't been. But I'd never expect it to be accurate on
>> those.
>>
> 	Well... the MaxMind demo of "my IP" did get the proper metropolitan
> area... But they list the ISP as "AT&T"... My real ISP is Earthlink
> (piggybacking on AT&T DSL service).
>
> 	The Lat/Long, however shows as
>
> 42.9634 -85.6681
> whereas a recent GPS readout shows
> 42.9159 -85.5541
>
> or 2m50s too far north, and 6m50s too far west.
>
> 	Same website, accessed from my Blackberry phone, gave a result of
> "United States, NA" and location 38 -97
>


I have read all your answer very carefully but i still need some way of 
getting it done.

All my Greek website visitors say they are from Europe/Athens which is 
the ISP's location and not user's homeland.

Well it worked for me but as many other told me it wasn't accurate for 
them too.

Please try this:  http://www.maxmind.com/en/geoip_demo

and tell me if maxmind's database can pippont you city's location.

Thank you.

-- 
What is now proved was at first only imagined!

[toc] | [prev] | [next] | [standalone]

#50573

On Sat, Jul 13, 2013 at 3:48 PM, Νικόλας <nikos@superhost.gr> wrote:
> Στις 13/7/2013 2:04 πμ, ο/η Dennis Lee Bieber έγραψε:
>>
>> On Sat, 13 Jul 2013 02:47:38 +1000, Chris Angelico <rosuav@gmail.com>
>> declaimed the following:
>>
>>>
>>> Oh, and just for laughs, I tried a few of my recent mobile IP
>>> addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
>>> some in the CBD and some out in the suburbs, but all vastly wrong, and
>>> places I haven't been. But I'd never expect it to be accurate on
>>> those.
>>>
>>         Well... the MaxMind demo of "my IP" did get the proper
>> metropolitan
>> area... But they list the ISP as "AT&T"... My real ISP is Earthlink
>> (piggybacking on AT&T DSL service).
>>
>>         The Lat/Long, however shows as
>>
>> 42.9634 -85.6681
>> whereas a recent GPS readout shows
>> 42.9159 -85.5541
>>
>> or 2m50s too far north, and 6m50s too far west.
>>
>>         Same website, accessed from my Blackberry phone, gave a result of
>> "United States, NA" and location 38 -97
>>
>
>
> I have read all your answer very carefully but i still need some way of
> getting it done.
>
> All my Greek website visitors say they are from Europe/Athens which is the
> ISP's location and not user's homeland.
>
> Well it worked for me but as many other told me it wasn't accurate for them
> too.
>
> Please try this:  http://www.maxmind.com/en/geoip_demo
>
> and tell me if maxmind's database can pippont you city's location.

Nikos, you keep asking for a way to do the impossible. We keep telling
you that it is impossible. No alternative technique will do what
cannot be done!

I just tried that on my two IPs and it was quite wrong on both of them
- further wrong than some of the others have been.

Stop expecting magic.

ChrisA

[toc] | [prev] | [next] | [standalone]

#50574

Στις 13/7/2013 8:53 πμ, ο/η Chris Angelico έγραψε:
> On Sat, Jul 13, 2013 at 3:48 PM, ������� <nikos@superhost.gr> wrote:
>> ���� 13/7/2013 2:04 ��, �/� Dennis Lee Bieber ������:
>>>
>>> On Sat, 13 Jul 2013 02:47:38 +1000, Chris Angelico <rosuav@gmail.com>
>>> declaimed the following:
>>>
>>>>
>>>> Oh, and just for laughs, I tried a few of my recent mobile IP
>>>> addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
>>>> some in the CBD and some out in the suburbs, but all vastly wrong, and
>>>> places I haven't been. But I'd never expect it to be accurate on
>>>> those.
>>>>
>>>          Well... the MaxMind demo of "my IP" did get the proper
>>> metropolitan
>>> area... But they list the ISP as "AT&T"... My real ISP is Earthlink
>>> (piggybacking on AT&T DSL service).
>>>
>>>          The Lat/Long, however shows as
>>>
>>> 42.9634 -85.6681
>>> whereas a recent GPS readout shows
>>> 42.9159 -85.5541
>>>
>>> or 2m50s too far north, and 6m50s too far west.
>>>
>>>          Same website, accessed from my Blackberry phone, gave a result of
>>> "United States, NA" and location 38 -97
>>>
>>
>>
>> I have read all your answer very carefully but i still need some way of
>> getting it done.
>>
>> All my Greek website visitors say they are from Europe/Athens which is the
>> ISP's location and not user's homeland.
>>
>> Well it worked for me but as many other told me it wasn't accurate for them
>> too.
>>
>> Please try this:  http://www.maxmind.com/en/geoip_demo
>>
>> and tell me if maxmind's database can pippont you city's location.
>
> Nikos, you keep asking for a way to do the impossible. We keep telling
> you that it is impossible. No alternative technique will do what
> cannot be done!
>
> I just tried that on my two IPs and it was quite wrong on both of them
> - further wrong than some of the others have been.
>
> Stop expecting magic.
>
> ChrisA
>

But it works for me, How can it be impossible and worked for me at the 
same time?

Also i tried some other website that asked me to allow it to run a 
javascript on my browser and it pinpointed even my street!

If it wasnt possbile then MaxMind would be seeling its GeoIP2 app for 
1380$ per year.

-- 
What is now proved was at first only imagined!

[toc] | [prev] | [next] | [standalone]

#50575

On Sat, Jul 13, 2013 at 4:07 PM, Νικόλας <nikos@superhost.gr> wrote:
> But it works for me, How can it be impossible and worked for me at the same
> time?

If I roll ten six-sided dice, will they total 35? Maybe. Maybe they'll
be close. But it's impossible to come up with a table for rolling
those dice on that will guarantee you exactly 35 every time.

ChrisA

[toc] | [prev] | [next] | [standalone]

#50576

Στις 13/7/2013 9:22 πμ, ο/η Chris Angelico έγραψε:
> On Sat, Jul 13, 2013 at 4:07 PM, ������� <nikos@superhost.gr> wrote:
>> But it works for me, How can it be impossible and worked for me at the same
>> time?
>
> If I roll ten six-sided dice, will they total 35? Maybe. Maybe they'll
> be close. But it's impossible to come up with a table for rolling
> those dice on that will guarantee you exactly 35 every time.


I just had some other friends of me, who live in different cities around 
Greece to test the link i gave to you in my previous post and for all of 
them it returned the correct city of their origin.

Seems like GeoIP2 is doing a better job that its predecessor GeopIP.

-- 
What is now proved was at first only imagined!

[toc] | [prev] | [next] | [standalone]

#50583

Νικόλας <nikos@superhost.gr> writes:

> But it works for me, How can it be impossible and worked for me at the
> same time?

Read the answers you got. What is *impossible* is *exactly and
precisely* find the geographical location of your machine, just using an
IP database like what you are doing, i.e. without your local machine
cooperating in the process (for example, by providing the numbers taken
from your local machine's GPS device).

What you are using is just some kind of heuristic, it's not an exact
science: each ISP is generally assigned a bunch of IPs which it manages
in some way, reserving some of them as *static IP* (that is, the same IP
is assigned to the same contractor, always), and assigning the other on
demand, like a DHCP server would do in a intranet. Think to the latter:
do you think it is possible to exactly locate in which room every
wireless notebook that travels inside your big house is at any given
time? Hint: no, you cannot, the best you can say is that each notebook
may be within a “sphere” of radius 50mt (just making up a number, it
obviously depend on the wireless signal power, and eventually on the
presence of wireless repeaters...).

Maybe you did provide your exact street address when signed the contract
with your ISP, but now ask yourself: would you be happy if your ISP
gives that kind of information to whomever may ask for it (in the
specific case, a geolocation service like maxmind.com)?

ciao, lele.
--
nickname: Lele Gaifax | Quando vivrò di quello che ho pensato ieri
real: Emanuele Gaifas | comincerò ad aver paura di chi mi copia.
lele@metapensiero.it  |                 -- Fortunato Depero, 1929.

[toc] | [prev] | [next] | [standalone]

#50598

On Sat, 13 Jul 2013 10:12:23 +0200, Lele Gaifax <lele@metapensiero.it>
declaimed the following:

>
>Read the answers you got. What is *impossible* is *exactly and
>precisely* find the geographical location of your machine, just using an
>IP database like what you are doing, i.e. without your local machine
>cooperating in the process (for example, by providing the numbers taken
>from your local machine's GPS device).
>
	Which obviously did NOT happen when I used my Blackberry phone -- which
does have geolocation enabled (it is tagging photos with the city).

	The only thing MaxMind had available from my phone was apparently the
DHCP server used by Blackberry, and gave out the approximate center of the
US.
-- 
	Wulfraed                 Dennis Lee Bieber         AF6VN
    wlfraed@ix.netcom.com    HTTP://wlfraed.home.netcom.com/

[toc] | [prev] | [next] | [standalone]

#50600

In article <krqqqt$eol$4@news.grnet.gr>,
 ÉΪÉ«É»όλας <nikos@superhost.gr> wrote:

> But it works for me, How can it be impossible and worked for me at the 
> same time?
> 
> Also i tried some other website that asked me to allow it to run a 
> javascript on my browser and it pinpointed even my street!
> 
> If it wasnt possbile then MaxMind would be seeling its GeoIP2 app for 
> 1380$ per year.

At Songza, we purchase the MaxMind database for doing geolocation based 
on IP address.  This is how we enforce our content licenses which only 
allow us to stream music to the US and Canada.  We also use it to target 
specific features (or advertising) to specific cities or other 
geographic areas within the US and Canada.

That being said, we understand that it is only an approximate solution.  
It is useful, but not perfect.  If you go to the MaxMind web site, you 
will see they have a range of products, at different prices, which 
promise various levels of accuracy and error rates.  But none of them, 
for any amount of money, offers "resolution down to the street address 
and 0% error rate".

[toc] | [prev] | [next] | [standalone]

#50610

[Multipart message — attachments visible in raw view] — view raw

On Sat, 13 Jul 2013, Νικόλας wrote:
> But it works for me, How can it be impossible and worked for me at the 
> same time?

2 + 2 = 4
2 + 6 = 8???

Why can't I make 2 and 6 equal 4? It worked for 2, so I know it's not
impossible! I don't care what everyone says, I was able to make one case work
so obviously I juat need to figure out how to make it work!


Allegorically,
W

[toc] | [prev] | [next] | [standalone]

#50566

On Sat, Jul 13, 2013 at 9:04 AM, Dennis Lee Bieber
<wlfraed@ix.netcom.com> wrote:
> On Sat, 13 Jul 2013 02:47:38 +1000, Chris Angelico <rosuav@gmail.com>
> declaimed the following:
>
>>
>>Oh, and just for laughs, I tried a few of my recent mobile IP
>>addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
>>some in the CBD and some out in the suburbs, but all vastly wrong, and
>>places I haven't been. But I'd never expect it to be accurate on
>>those.
>>
>         Well... the MaxMind demo of "my IP" did get the proper metropolitan
> area... But they list the ISP as "AT&T"... My real ISP is Earthlink
> (piggybacking on AT&T DSL service).
>
>         The Lat/Long, however shows as
>
> 42.9634 -85.6681
> whereas a recent GPS readout shows
> 42.9159 -85.5541
>
> or 2m50s too far north, and 6m50s too far west.
>
>         Same website, accessed from my Blackberry phone, gave a result of
> "United States, NA" and location 38 -97

When you try to place a visitor geographically by IP address, the only
thing you can be absolutely 100% certain of is which RIR they're at
(proxies aside - you're just testing the proxy rather than the
ultimate origin). Country is also highly likely to be right, though
not certain (I've never known it to be wrong, but I've never been able
to confirm what happens with some of the small European countries -
for all I know they could share ISPs and netblocks). Anything tighter
than that is goign to be pretty hit-and-miss. But I have to say, it's
improved a lot over the years. Back in the early 2000s - say, about 8
years ago, I think - I was playing with this sort of technology, and
it placed me in Sydney. That's one state away, lots of rivalry
separating us (friendly rivalry, of course; in a country that's doing
its best to kill us all, we can't afford to really hate each other),
and roughly 750-1000km wrong by distance (depending on how you measure
- most people don't put an odometer on a crow). So at least now it
gets within the same degree of latitude and longitude... most of the
time.

ChrisA

[toc] | [prev] | [next] | [standalone]

#50685

[Multipart message — attachments visible in raw view] — view raw

On Fri, Jul 12, 2013 at 7:04 PM, Dennis Lee Bieber <wlfraed@ix.netcom.com>wrote:

> On Sat, 13 Jul 2013 02:47:38 +1000, Chris Angelico <rosuav@gmail.com>
> declaimed the following:
>
> >
> >Oh, and just for laughs, I tried a few of my recent mobile IP
> >addresses in the GeoIP lookup. All of them quoted Melbourne someplace,
> >some in the CBD and some out in the suburbs, but all vastly wrong, and
> >places I haven't been. But I'd never expect it to be accurate on
> >those.
> >
>         Well... the MaxMind demo of "my IP" did get the proper metropolitan
> area... But they list the ISP as "AT&T"... My real ISP is Earthlink
> (piggybacking on AT&T DSL service).
>
>         The Lat/Long, however shows as
>
> 42.9634 -85.6681
> whereas a recent GPS readout shows
> 42.9159 -85.5541
>
> or 2m50s too far north, and 6m50s too far west.
>
>         Same website, accessed from my Blackberry phone, gave a result of
> "United States, NA" and location 38 -97
> --
>         Wulfraed                 Dennis Lee Bieber         AF6VN
>     wlfraed@ix.netcom.com    HTTP://wlfraed.home.netcom.com/
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>

Speaking more from a political perspective, an important aspect of the
internet is that if you are a publisher who doesn't require registration to
read what you post, your readers are free to be more or less anonymous.
This is a good thing in many ways.  If you have a service that requires
some sort of sign up, then you can require knowing things about them
(location, etc).  If you want to know where your readers are, you need to
make arrangements with their ISPs to get that information, since in many
cases the ISP has a physical link to your machine.  But why should they
provide that to you?  After all you are not their customer.

It might be fun to know, but the repercussions are serious.


-- 
Joel Goldstick
http://joelgoldstick.com

[toc] | [prev] | [next] | [standalone]

#50549

On Fri, Jul 12, 2013 at 10:32 AM, Νικόλας <nikos@superhost.gr> wrote:
>
> I know i have asked before but hwta i get is ISP city not visitors precise
> city.
>
> GeoLiteCity.dat isnt accurate that's why it comes for free.
> i must somehow get access to GeoIPCity.dat which is the full version.
>
> And of course it can be done, i dont want to believe that it cant.
>
> When visiting http://www.geoiptool.com/en/__ip_info/ it pinpoints my _exact_
> city of living, not the ISP's.
> It did not even ask me to allow a geop ip javascript to run it present sit
> instantly.

Try this:

1) Go to http://incloak.com (or any other free web proxy site).
2) Paste in the URL http://www.geoiptool.com and press Enter
3) See where it thinks you are now.

When I tried it, it placed me on the wrong side of the Atlantic Ocean.

[toc] | [prev] | [next] | [standalone]

#50555

On 7/12/2013 1:19 PM, Ian Kelly wrote:

> Try this:
>
> 1) Go to http://incloak.com (or any other free web proxy site).
> 2) Paste in the URL http://www.geoiptool.com and press Enter
> 3) See where it thinks you are now.
>
> When I tried it, it placed me on the wrong side of the Atlantic Ocean.

Me to. Thanks for the link.

-- 
Terry Jan Reedy

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | comp.lang.python

csiph-web