Sharad,
Could you share with the rest of us?

Thanks,
Justin

On Wednesday, August 6, 2014 2:36:48 PM UTC-4, Sharad Lele wrote:
>
>  Dear Justin:
>
> I beg to disagree. For Karnataka, which I know best, GADM has 176 
> features. Census 2001 shows 175 talukas. Census 2011 has 176 unique taluka 
> codes.
>
> Second, boundaries in Karnataka of GADM are okay: i am comparing them with 
> taluka boundaries I have digitzed from Census handbook itself.
>
> Sharad
>
> On 06-Aug-14 11:42 PM, Justin Meyers wrote:
>  
> Sharad, 
>
>
> <https://lh5.googleusercontent.com/-45mYgnZbcC0/U-Ju6KmGRCI/AAAAAAAAAKQ/FNF491-pg_0/s1600/gadm.jpg>
> GADM is junk.  The person who created it doesn't respond to e-mails.  I 
> reached out to him (Dr. Robert Hijmans) a few years back and he was 
> useless.  Currently GADM has 2299 coundaries.  See a pic of the two layers 
> compared (GADM is Green, this dataset is red)
>
>  
>  Justin
>
>
> On Wednesday, August 6, 2014 1:58:20 PM UTC-4, Sharad Lele wrote: 
>>
>> How does this compare with the maps from the GADM database?
>>
>> My notes say GADM database maps (India state, district and taluka) show 
>> district boundaries of 2001 or earlier, with spatial positioning error of 
>> about 1km, and much more generalized than the boundaries shown on Survey of 
>> India toposheets. (goes without saying since the topos are at much finer 
>> scale than such maps)
>>
>> Sharad
>>
>> On Wednesday, August 6, 2014 10:49:42 AM UTC+5:30, Dilip Damle wrote: 
>>>
>>> Justin, Devdatta, 
>>>
>>> It's all about how things work in India, 
>>>
>>> Devdatta mentioned SWEat,blood and teARS 
>>>
>>> It also includes the capitalised word :) 
>>>
>>>  Good to persevere though.
>>> Congratulatioins.
>>>
>>>  Rgds
>>> Dilip Damle
>>>
>>>
>>>
>>>
>>> On Wednesday, August 6, 2014 9:59:13 AM UTC+5:30, Justin Meyers wrote: 
>>>>
>>>>  
>>>> <https://lh5.googleusercontent.com/-R0Y80NzLfCo/U-GuxQ7iCwI/AAAAAAAAAKA/gdN3wX4skFk/s1600/bihar_sample.jpg>
>>>> see the Bihar data compared to this:Red=Bihar Black=this dataset 
>>>>
>>>>
>>>>
>>>> On Wednesday, August 6, 2014 12:21:27 AM UTC-4, Justin Meyers wrote: 
>>>>>
>>>>> Devdatta, 
>>>>> This is raw data that I converted from kml to shp.  So it took me 
>>>>> about 30 minutes to make sure the attributes came through properly.  
>>>>> Cannot 
>>>>> say my source until I download the rest of the content from the site, 
>>>>> before they take it down.  It seems like they put data on the internet, 
>>>>> then try to take it down (or hide it...).  Not really sure why?  I posted 
>>>>> the Orissa sample for you some time ago, but you never responded?  I 
>>>>> found 
>>>>> official data for Orissa as well, so converting the census pdfs to gis is 
>>>>> a 
>>>>> big waste of time (took me a few days...).
>>>>>
>>>>>  As per the quality of this dataset, it is the best I have seen on a 
>>>>> national level.  But if you look state by state, or really want quality 
>>>>> data, then this isn't what you want.  I would call this dataset a rough 
>>>>> start.
>>>>>
>>>>>  I posted a update about Bihar (
>>>>> https://groups.google.com/forum/#!topic/datameet/QuP4Tt4n0_g) and 
>>>>> that data is amazing, this data doesn't hold a candle to it.  Too bad the 
>>>>> government doesn't want help from the experts ;)
>>>>>
>>>>>  Cheers,
>>>>> Justin
>>>>>
>>>>>  
>>>>>
>>>>> On Wednesday, August 6, 2014 12:02:57 AM UTC-4, Devdatta Tengshe 
>>>>> wrote: 
>>>>>>
>>>>>>    Hi Justin,
>>>>>>
>>>>>>  Firstly let me congratulate you on doing this. I'm sure that this 
>>>>>> must have involved a lot of sweat, blood and tears (Working with 
>>>>>> government 
>>>>>> data, usually involves the last one). This is the most up-to-date, 
>>>>>> comprehensive & complete dataset for sub districts that I have seen.
>>>>>>
>>>>>>  Could you tell us what was the source for creating these boundaries?
>>>>>>
>>>>>>  I'll cross check it with other datasets, and let you know how it 
>>>>>> goes.
>>>>>>
>>>>>>  Regards,
>>>>>>  Devdatta
>>>>>>   
>>>>>>   
>>>>>>
>>>>>> On Wed, Aug 6, 2014 at 8:06 AM, Justin Meyers <[email protected]
>>>>>> > wrote:
>>>>>>
>>>>>>>  
>>>>>>> <https://lh5.googleusercontent.com/-JCJJdFxwIJ4/U-GSaLtnplI/AAAAAAAAAJw/3nhGoIFzCNE/s1600/indiatahsil2001.jpg>
>>>>>>>
>>>>>>> https://app.box.com/s/vljeiolbqhx2hwyz8qxs
>>>>>>>  
>>>>>>>  Processed from a kml; did not edit any geometry or attributes. 
>>>>>>>  Data is pretty generalized, but it is a good start for a national 
>>>>>>> dataset. 
>>>>>>>  I keep finding better data as I continue to look (already posted two 
>>>>>>> previous versions of a taluk/ tahsil/ amin 3 boundary).  After concerns 
>>>>>>> to 
>>>>>>> missing data, this is the best to date I have found.
>>>>>>>
>>>>>>>  Cheers!
>>>>>>>  Justin
>>>>>>>  -- 
>>>>>>> Datameet is a community of Data Science enthusiasts in India. Know 
>>>>>>> more about us by visiting http://datameet.org
>>>>>>> --- 
>>>>>>> You received this message because you are subscribed to the Google 
>>>>>>> Groups "datameet" group.
>>>>>>> To unsubscribe from this group and stop receiving emails from it, 
>>>>>>> send an email to [email protected].
>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>>
>>>>>>  
>>>>>>         -- 
> Datameet is a community of Data Science enthusiasts in India. Know more 
> about us by visiting http://datameet.org
> --- 
> You received this message because you are subscribed to a topic in the 
> Google Groups "datameet" group.
> To unsubscribe from this topic, visit 
> https://groups.google.com/d/topic/datameet/GJKMXXSB-Ns/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to 
> [email protected] <javascript:>.
> For more options, visit https://groups.google.com/d/optout.
>
>
> -- 
> Democratizing Forest Governance in India
> (In press with Oxford University Press India)
>
>  

-- 
Datameet is a community of Data Science enthusiasts in India. Know more about 
us by visiting http://datameet.org
--- 
You received this message because you are subscribed to the Google Groups 
"datameet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to