Most of my last works on Amara now are not about xml processing but related to html (web scraping). Other libs are doing more helpers about parsing html (lxml with cssselects, iterlinks, sourceline, ...) and are other aproaches to extract info (pyparsing) and to access it (mechanize, selenium). What is your experience?
Should we let amara as it is now, a very good generic xml parser and use other libs for more specialized works? Or transform Amara into a Big Data ready lib in order to extract / analize / visualize unstructured data?
On 22 April 2012 10:07, Luis Miguel Morillas <moril...@gmail.com> wrote:
> Hi all,
> Most of my last works on Amara now are not about xml processing but > related to html (web scraping). Other libs are doing more helpers > about parsing html (lxml with cssselects, iterlinks, sourceline, ...) > and are other aproaches to extract info (pyparsing) and to access it > (mechanize, selenium). What is your experience?
> Should we let amara as it is now, a very good generic xml parser and > use other libs for more specialized works? Or transform Amara into a > Big Data ready lib in order to extract / analize / visualize > unstructured data?
My only use has been to generate XML, it was the most sensible Python library I found. One mans view.
Before responding to Luis's question, I would like to echo something said by Sylvain in the "What is the status of Amara?" thread: it's quite difficult to convince management that Amara is the right tool for the job when it's been in alpha since January 2009. To respond to Luis's concerns about Amara's uptake, I feel reasonably certain that other potential users and contributors are also hurt by the dynamic outlined above. I know we would have used another product if it did a similar job and wasn't in alpha.
From this experience, I strongly encourage providing stable Amara releases more often. If I could be so bold, I think that planning these releases on a product roadmap would be a great way to accomplish this. Following Luis's comments, I could imagine that Amara 2.0 will include the core engine and other stable code while Amara 2.1 would include these HTML libraries, and so on. Not only would this roadmap advertise what the product can and will be able to do, it will also encourage community feedback about where the product should go and give users stable software with which to start providing feedback, tests, and bugfixes in order to enhance the functionality.
Amara is an absolutely phenomenal tool and clearly fills a gap that is not filled by any other XML tool I have seen. It deserves to have a huge community behind it, and with a little bit of website and process tweaking, this project really could take off.
-----Original Message-----
From: akara@googlegroups.com [mailto:akara@googlegroups.com] On Behalf Of Dave Pawson
Sent: Sunday, April 22, 2012 5:42 AM
To: akara@googlegroups.com
Subject: Re: The future of amara / akara project: 2.0 stable or 3.0 ?
On 22 April 2012 10:07, Luis Miguel Morillas <moril...@gmail.com> wrote:
> Hi all,
> Most of my last works on Amara now are not about xml processing but
> related to html (web scraping). Other libs are doing more helpers
> about parsing html (lxml with cssselects, iterlinks, sourceline, ...)
> and are other aproaches to extract info (pyparsing) and to access it
> (mechanize, selenium). What is your experience?
> Should we let amara as it is now, a very good generic xml parser and
> use other libs for more specialized works? Or transform Amara into a
> Big Data ready lib in order to extract / analize / visualize
> unstructured data?
My only use has been to generate XML, it was the most sensible Python library I found.
One mans view.
--
You received this message because you are subscribed to the Google Groups "akara" group.
To post to this group, send email to akara@googlegroups.com.
To unsubscribe from this group, send email to akara+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/akara?hl=en.
> Before responding to Luis's question, I would like to echo something said > by Sylvain in the "What is the status of Amara?" thread: it's quite > difficult to convince management that Amara is the right tool for the job > when it's been in alpha since January 2009. To respond to Luis's concerns > about Amara's uptake, I feel reasonably certain that other potential users > and contributors are also hurt by the dynamic outlined above. I know we > would have used another product if it did a similar job and wasn't in alpha.
> From this experience, I strongly encourage providing stable Amara releases > more often. If I could be so bold, I think that planning these releases on > a product roadmap would be a great way to accomplish this. Following > Luis's comments, I could imagine that Amara 2.0 will include the core > engine and other stable code while Amara 2.1 would include these HTML > libraries, and so on. Not only would this roadmap advertise what the > product can and will be able to do, it will also encourage community > feedback about where the product should go and give users stable software > with which to start providing feedback, tests, and bugfixes in order to > enhance the functionality.
> Amara is an absolutely phenomenal tool and clearly fills a gap that is not > filled by any other XML tool I have seen. It deserves to have a huge > community behind it, and with a little bit of website and process tweaking, > this project really could take off.
You're certainly right about all that, and all the stuff you mention is stuff that I've always wanted to do, but I have so much going on that contemplating it all is nearly hopeless. I do appreciate that folks such as Luis and Werner are really digging in to help, because that puts a far more direct sort of pressure on me to find the time to support them. The general idea of growing Amara's community is a nice one, but the simple, awkward fact is that such an idea alone isn't quite enough for me. I use Amara all the time in work (nice to be my own manager ;) ), and it generally gives me an edge in getting stuff done, so I find myself needing additional motivation to push it to the next level in order to make *other* managers happy.
I guess my point in saying the above is that maybe the current flurry of activity will help. Certainly I don't want to promise any more than a specific focus at a time. Test and docs are the biggest issue with Amara right now, and incidentally, if we can get tests and docs much improved, I think we can easily make it beta status. And then maybe we can look at the next in line of the problems you mention.
On Tue, Apr 24, 2012 at 2:40 PM, Uche Ogbuji <u...@ogbuji.net> wrote: > On Mon, Apr 23, 2012 at 11:56 AM, Pigneri, Rocco <RPign...@lavastorm.com>wrote:
>> Dear all,
>> Before responding to Luis's question, I would like to echo something said >> by Sylvain in the "What is the status of Amara?" thread: it's quite >> difficult to convince management that Amara is the right tool for the job >> when it's been in alpha since January 2009. To respond to Luis's concerns >> about Amara's uptake, I feel reasonably certain that other potential users >> and contributors are also hurt by the dynamic outlined above. I know we >> would have used another product if it did a similar job and wasn't in alpha.
>> From this experience, I strongly encourage providing stable Amara >> releases more often. If I could be so bold, I think that planning these >> releases on a product roadmap would be a great way to accomplish this. >> Following Luis's comments, I could imagine that Amara 2.0 will include the >> core engine and other stable code while Amara 2.1 would include these HTML >> libraries, and so on. Not only would this roadmap advertise what the >> product can and will be able to do, it will also encourage community >> feedback about where the product should go and give users stable software >> with which to start providing feedback, tests, and bugfixes in order to >> enhance the functionality.
>> Amara is an absolutely phenomenal tool and clearly fills a gap that is >> not filled by any other XML tool I have seen. It deserves to have a huge >> community behind it, and with a little bit of website and process tweaking, >> this project really could take off.
> You're certainly right about all that, and all the stuff you mention is > stuff that I've always wanted to do, but I have so much going on that > contemplating it all is nearly hopeless. I do appreciate that folks such as > Luis and Werner are really digging in to help, because that puts a far more > direct sort of pressure on me to find the time to support them. The general > idea of growing Amara's community is a nice one, but the simple, awkward > fact is that such an idea alone isn't quite enough for me. I use Amara all > the time in work (nice to be my own manager ;) ), and it generally gives me > an edge in getting stuff done, so I find myself needing additional > motivation to push it to the next level in order to make *other* managers > happy.
> I guess my point in saying the above is that maybe the current flurry of > activity will help. Certainly I don't want to promise any more than a > specific focus at a time. Test and docs are the biggest issue with Amara > right now, and incidentally, if we can get tests and docs much improved, I > think we can easily make it beta status. And then maybe we can look at the > next in line of the problems you mention.
I can definitely relate to the fact that people's expectations isn't enough to motivate oneself at spending the little time one has. I can also understand why you'd rather have better tests and docs before moving to beta. But, if you use it on a daily basis, doesn't it mean it's already worth a beta? I know you'd rather have tests that prove it but aren't you afraid that waiting for them to be written, by whomever, is a sure way of never reaching that beta status?
Maybe Amara has simply reached a "Production/Stable" status already which would explain why there's hardly any omg-it-even-bakes-eggs-kind-of-buzz over it, but I assume Amara isn't cheerleading anyhow. We have reached a similar point with CherryPy and it's a tough one. What we've tried in the recent months was to make the doc shorter, more to the point and clearer. In addition, I've started writing simple recipes to point at when people have common questions. It's not much but it has helped without making people too frustrated. Just as an example...
> On Tue, Apr 24, 2012 at 2:40 PM, Uche Ogbuji <u...@ogbuji.net> wrote:
>> On Mon, Apr 23, 2012 at 11:56 AM, Pigneri, Rocco <RPign...@lavastorm.com>
>> wrote:
>>> Dear all,
>>> Before responding to Luis's question, I would like to echo something said
>>> by Sylvain in the "What is the status of Amara?" thread: it's quite
>>> difficult to convince management that Amara is the right tool for the job
>>> when it's been in alpha since January 2009. To respond to Luis's concerns
>>> about Amara's uptake, I feel reasonably certain that other potential users
>>> and contributors are also hurt by the dynamic outlined above. I know we
>>> would have used another product if it did a similar job and wasn't in alpha.
>>> From this experience, I strongly encourage providing stable Amara
>>> releases more often. If I could be so bold, I think that planning these
>>> releases on a product roadmap would be a great way to accomplish this.
>>> Following Luis's comments, I could imagine that Amara 2.0 will include the
>>> core engine and other stable code while Amara 2.1 would include these HTML
>>> libraries, and so on. Not only would this roadmap advertise what the
>>> product can and will be able to do, it will also encourage community
>>> feedback about where the product should go and give users stable software
>>> with which to start providing feedback, tests, and bugfixes in order to
>>> enhance the functionality.
>>> Amara is an absolutely phenomenal tool and clearly fills a gap that is
>>> not filled by any other XML tool I have seen. It deserves to have a huge
>>> community behind it, and with a little bit of website and process tweaking,
>>> this project really could take off.
>> You're certainly right about all that, and all the stuff you mention is
>> stuff that I've always wanted to do, but I have so much going on that
>> contemplating it all is nearly hopeless. I do appreciate that folks such as
>> Luis and Werner are really digging in to help, because that puts a far more
>> direct sort of pressure on me to find the time to support them. The general
>> idea of growing Amara's community is a nice one, but the simple, awkward
>> fact is that such an idea alone isn't quite enough for me. I use Amara all
>> the time in work (nice to be my own manager ;) ), and it generally gives me
>> an edge in getting stuff done, so I find myself needing additional
>> motivation to push it to the next level in order to make *other* managers
>> happy.
>> I guess my point in saying the above is that maybe the current flurry of
>> activity will help. Certainly I don't want to promise any more than a
>> specific focus at a time. Test and docs are the biggest issue with Amara
>> right now, and incidentally, if we can get tests and docs much improved, I
>> think we can easily make it beta status. And then maybe we can look at the
>> next in line of the problems you mention.
> I can definitely relate to the fact that people's expectations isn't enough
> to motivate oneself at spending the little time one has. I can also
> understand why you'd rather have better tests and docs before moving to
> beta. But, if you use it on a daily basis, doesn't it mean it's already
> worth a beta? I know you'd rather have tests that prove it but aren't you
> afraid that waiting for them to be written, by whomever, is a sure way of
> never reaching that beta status?
> Maybe Amara has simply reached a "Production/Stable" status already which
> would explain why there's hardly any omg-it-even-bakes-eggs-kind-of-buzz
> over it, but I assume Amara isn't cheerleading anyhow. We have reached a
> similar point with CherryPy and it's a tough one. What we've tried in the
> recent months was to make the doc shorter, more to the point and clearer. In
> addition, I've started writing simple recipes to point at when people have
> common questions. It's not much but it has helped without making people too
> frustrated. Just as an example...
I agree with Sylvain that Amara and Akara are not alpha software de
facto. They can be used in production systems without risk.
Uche, I don't want you have a bad sensation about your work on the
project. It's a nice and cool project. If Amara grows so slowly now
it's because the rest of the users/developers should contribute more,
not only you. Amara/akara are based in legacy projects, but Foursuite
doesn't exist now, and the group should coordinate better their little
works.
> --
> You received this message because you are subscribed to the Google Groups
> "akara" group.
> To post to this group, send email to akara@googlegroups.com.
> To unsubscribe from this group, send email to
> akara+unsubscribe@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/akara?hl=en.
