UWECProgrammer/ConsolidatedMentionResolutionBenchmark

sentence stringlengths 1 1.38k	label stringclasses 3 values
@Pelegi, can you mention just one thing that JBoss does better than Glassfish?	o
It's strictly what you're used to.	o
I found configuring and using JBoss to be about as pleasing as trying to shave with a jagged spoon while riding a unicycle with no seat.	o
Glassfish however, just seems to work.	o
My choice is Glassfish: it is faster to start-up, to deploy web applications.the Admin console is very accurate, moreover it is both graphical and command-line.Now it is fully-supported by Oracle, its future is not uncertain any more.It is based upon OSGi: it is really modular.You can install Spring-osgi applications easly.You can freely choice between Eclipse Link , Apache Open JPA and Hibernate without pain.It is greatly integrated in NetBeans (and, of course, Eclipse).	p
On the other hand, Jboss is more widespread and thus it is easy to find job on it (IMHO because it is an old App Server; glassfish is relatively new).	p
I started developing a Seam application on JBoss and after a few months tried to get it working on Glassfish.	p
I found the Glassfish enviroment very counterintuitive.	o
I had to fix some strange errors on GF and I could not find any information on the web so after a few days I gave up and went back to JBoss.	o
So personally I would not recommend Glassfish.	n
I will have to go with Glassfish on this.	o
Deploying a diverse range of application built on different webservices, we had major compatibility issues, resulting in customizing the class loading & still faced many issues.	o
Glassfish I never faced any such issues.	o
Merits of Java applet for web applications.	p
What are the advantages and disadvantages of using a Java applet for authentication as part of a web application?	o
I am think of cases where such an applet is only used for authentication and transaction-signing purpose and not otherwise in the web application.	o
What would be the point of using an applet for authentication & signing?	o
Security theatre?	o
Most servers have built-in SPNEGO support, so you shouldn't have to code something like an applet to do your authentication.	o
Do you need to use any phisical device during authentication.	o
Such as hard drive, flash-card, smart-card, eToken, iKey, etc.?	o
In that case you must use the applet.	o
Yes, I understand the value when using h/w crypto devices but this is not the general case, including shared-secret and non-machine-readable h/w OTP devices.	o
As a best practice, an applet should only augment the server's capabilities in the context of authentication.	p
The server ought to re-verify what has been sent by the applet (client), for the message could have been modified in transit.	o
There are times however where an applet may serve a valid purpose.	o
For instance, messages sent by the client, may be digitally signed by the applet, with the keys being managed in per-user USB drives (containing the private keys).	o
In such a context, the applet serves as a mechanism for accessing the JCA libraries in Java, apart from accessing the USB store.	o
The signature produced will have to re-validated by the server on receipt of the message.	o
When implemented correctly, applets like any client-side security scheme will increase the security posture of an application.	o
When implemented incorrectly, it results in a security theatre where the server relies on the client to provide trusted input.	o
Also, if one needs to perform privileged operations (like access drives containing private keys), then the applet must be signed; and an appropriate Java security policy file must be maintained.	p
Any errors in the management process of the Java security model will obviously lower the security of the system.	o
Also, if native OS specific APIs like MSCAPI have to be accessed, then deployment of native code (DLLs) is a pre-requisite, and the software (not the applet) responsible for doing so, must be coded securely.	o
Disadvantages: Client JVM management.	n
Session/Proxy coordination management.	o
Antiquated technology.	n
Advantages: Single language skills?	o
...?	o
IMHO if you don't need something like powerful GUI support or Java2D (most of which can be done with JQuery, ExtJS, etc) there is no real argument for applets.	p
How to authenticate users from a rest service.	o
I am trying to authenticate users with a REST service I built using drop wizard.	o
From previous questions I found great example of authenticating with openID on github: URL_https://github.com/gary-rowe/DropwizardOpenID [ URL_https://github.com/gary-rowe/DropwizardOpenID ] However, I don't want to deal with openID at the moment and simply want users to 1.	p
Signup, 2.	o
Signin My questions/confusions are: 1.	n
For Signup: I'm thinking about sending users's username/password as a CODETERM1 request with the credentials as either form parameters or part of JSON body.	o
However, isn't there a security risk here of sending password in plain text?	o
2.	o
For Sing-in I'm thinking about using CODETERM2 in Dropwizard.	o
3.	o
I don't want to store passwords in plain text.	o
What strategy should I follow after I get the users' password in the CODETERM3 as plain text?	o
I'm looking for some java libraries that can assist in password salt and MD5 .	o
Have you read their [authentication]( URL_http://dropwizard.codahale.com/manual/auth /#oauth2) documentation?	o
OAuth2 is going to be the most secure and RESTful approach; as for sending creds in plaintext, are you building a web front-end?	o
If so, serve your site over https and it won't be a problem.	o
I don't envision any external apps using the API for data.	o
My api will just be used by my app's mobile and web interface.	o
Wouldn't Oauth2 be an overdo for this?	o
OAuth2 is a great approach to ensure that each logged in user receives their own access token that you can track and manage.	p
Short of dropping the API altogether (and using just session/cookies) it's still the best and most scalable approach.	p
Do the work now, it'll last (forever?	o
).	o
Maybe I'm not following correctly.	p
If I go down OAuth2 route I have to register my app with one of the OAuth2 providers e.g.	o
google, github, etc.	o
right?	o
And then I can use that token to authorize the user?	o
No no no... Dropwizard supports OAuth2 on the back end - so you would implement your own OAuth2 service, and then have the browser perform the auth handshake with it; completely independent from Google, et al.	o
Ah, roger that.	o
much better.	o
I'll find more info on how to implement OAuth2 with drop wizard.	o
Users can "signup" with Oauth2 as well?	o
Yes they can.	o
Looks like their docs are pretty good too, and being open source, you should find lots of examples.	p
Looking at the docs, we can see that Dropwizard supports a standalone OAuth2 implementation: URL_http://dropwizard.codahale.com/manual/auth/#oauth2 [ URL_http://dropwizard.codahale.com/manual/auth /#oauth2] OAuth2 has several advantages, many of which can be read about here: URL_http://stackoverflow.com/questions/7561631/oauth-2-0-benefits-and-use- cases-why [OAuth-2.0:-Benefits-and-use-cases-why?]	p
Things to note: when dealing with authentication, you should always host over HTTPS to ensure transport encryption Dropwizard claims their OAuth2 implementation isn't yet finalized, and may change in the future.	o
As a fall back, they do support Basic auth as well, which when used over HTTPS would be still reasonably secure.	p
Implementing this does not involve using any third party "social" authentication services such as Google or Facebook.	o
Thanks for the shout out for the Dropwizard OpenID project.	o
Glad it was able to get you started.	p
If you want a pure web form type approach, take a look at another of my projects URL_https://github.com/gary-rowe/MultiBitMerchant [MultiBit-Merchant] which provides multiple authentication methods (web form, HMAC, cookie).	o
You'll need to dig around to really see it working since this project is not designed as a demo as such and is very much a work in progress.	p
After loading the project, look for URL_https://github.com/gary- rowe/MultiBitMerchant/blob/develop/mbm-client/src/main/java/org/multibit/mbm/c lient/interfaces/rest/auth/webform/WebFormClientAuthenticator.java [CODETERM1] which will get you in the right area.	o
The general principles involved with Dropwizard authentication URL_http ://gary-rowe.com/agilestack/2012/10/23/multibit-merchant-implementing-hmac- authentication-in-dropwizard [are-discussed-in-this-blog-article] .	o
Although it targets HMAC you can easily adapt it for web form or cookie using the source code referenced earlier.	o
It's all MIT license so just use it as you need.	o
Thanks.	o
I am looking through the code and implementing my simple service in parallel.	p
I had one more question - URL_http://stackoverflow.com/questions/20662871/how-to-do-basic-authentication - of-a-resource-in-dropwizard.	o
Open Source Iris Scanning Library in Java.	p
I am looking for suggestions about open source Iris Scanning libraries in Java.	p
Has anyone used before which can built an unique profile for each scan & be able to authenticate too.	o
Thank you.	p
I haven't used Java libraries that "just do" that entire work you describe, but it's not that difficult to use ( URL_http://sourceforge.net/projects/javavis/ [ URL_http://sourceforge.net/projects/javavis/ ] ) as in the sample code described in the following paper: URL_http://imai-software.com/openlab/data/IMAI- JOUNALVI.pdf#page=43 [ URL_http://imai-software.com/openlab/data/IMAI - JOUNALVI.pdf#page=43] (see from page 43) - that paper is quite good and for it's length :).	n
Of course, there are a few "working" solutions, e.g.	p
: URL_http://mac.softpedia.com/get/Math-Scientific/JIRRM.shtml [ URL_http://mac.softpedia.com/get/Math-Scientific/JIRRM.shtml ] (Java based) URL_http://projectiris.co.uk/ [ URL_http://projectiris.co.uk/ ] (QT based -but there are more QT <-> Java bridges) but personally I haven't used in production (just for testing) these 2 libraries since the documentation is not very good/detailed, and since this kind of problem is quite "cutting edge" so the documentation and an explanation paper is much more important than any source code.	p
If none of the above approaches work for you, I would suggest to use a MATLAB based iris recognition examples (because there are many MATLAB based implementations, papers, theses) and than convert them to Java or use the directly from Java (since MATLAB has support for this).	o
Link to the paper is broken.	n
I would look into URL_http://opencv.willowgarage.com/wiki/ [OpenCV-which- is-a-C++-library-for-computer-vision] .	o

@Pelegi, can you mention just one thing that JBoss does better than Glassfish?

o

It's strictly what you're used to.

o

I found configuring and using JBoss to be about as pleasing as trying to shave with a jagged spoon while riding a unicycle with no seat.

o

Glassfish however, just seems to work.

o

My choice is Glassfish: it is faster to start-up, to deploy web applications.the Admin console is very accurate, moreover it is both graphical and command-line.Now it is fully-supported by Oracle, its future is not uncertain any more.It is based upon OSGi: it is really modular.You can install Spring-osgi applications easly.You can freely choice between Eclipse Link , Apache Open JPA and Hibernate without pain.It is greatly integrated in NetBeans (and, of course, Eclipse).

p

On the other hand, Jboss is more widespread and thus it is easy to find job on it (IMHO because it is an old App Server; glassfish is relatively new).

p

I started developing a Seam application on JBoss and after a few months tried to get it working on Glassfish.

p

I found the Glassfish enviroment very counterintuitive.

o

I had to fix some strange errors on GF and I could not find any information on the web so after a few days I gave up and went back to JBoss.

o

So personally I would not recommend Glassfish.

n

I will have to go with Glassfish on this.

o

Deploying a diverse range of application built on different webservices, we had major compatibility issues, resulting in customizing the class loading & still faced many issues.

o

Glassfish I never faced any such issues.

o

Merits of Java applet for web applications.

p

What are the advantages and disadvantages of using a Java applet for authentication as part of a web application?

o

I am think of cases where such an applet is only used for authentication and transaction-signing purpose and not otherwise in the web application.

o

What would be the point of using an applet for authentication & signing?

o

Security theatre?

o

Most servers have built-in SPNEGO support, so you shouldn't have to code something like an applet to do your authentication.

o

Do you need to use any phisical device during authentication.

o

Such as hard drive, flash-card, smart-card, eToken, iKey, etc.?

o

In that case you must use the applet.

o

Yes, I understand the value when using h/w crypto devices but this is not the general case, including shared-secret and non-machine-readable h/w OTP devices.

o

As a best practice, an applet should only augment the server's capabilities in the context of authentication.

p

The server ought to re-verify what has been sent by the applet (client), for the message could have been modified in transit.

o

There are times however where an applet may serve a valid purpose.

o

For instance, messages sent by the client, may be digitally signed by the applet, with the keys being managed in per-user USB drives (containing the private keys).

o

In such a context, the applet serves as a mechanism for accessing the JCA libraries in Java, apart from accessing the USB store.

o

The signature produced will have to re-validated by the server on receipt of the message.

o

When implemented correctly, applets like any client-side security scheme will increase the security posture of an application.

o

When implemented incorrectly, it results in a security theatre where the server relies on the client to provide trusted input.

o

Also, if one needs to perform privileged operations (like access drives containing private keys), then the applet must be signed; and an appropriate Java security policy file must be maintained.

p

Any errors in the management process of the Java security model will obviously lower the security of the system.

o

Also, if native OS specific APIs like MSCAPI have to be accessed, then deployment of native code (DLLs) is a pre-requisite, and the software (not the applet) responsible for doing so, must be coded securely.

o

Disadvantages: Client JVM management.

n

Session/Proxy coordination management.

o

Antiquated technology.

n

Advantages: Single language skills?

o

...?

o

IMHO if you don't need something like powerful GUI support or Java2D (most of which can be done with JQuery, ExtJS, etc) there is no real argument for applets.

p

How to authenticate users from a rest service.

o

I am trying to authenticate users with a REST service I built using drop wizard.

o

From previous questions I found great example of authenticating with openID on github: URL_https://github.com/gary-rowe/DropwizardOpenID [ URL_https://github.com/gary-rowe/DropwizardOpenID ] However, I don't want to deal with openID at the moment and simply want users to 1.

p

Signup, 2.

o

Signin My questions/confusions are: 1.

n

For Signup: I'm thinking about sending users's username/password as a CODETERM1 request with the credentials as either form parameters or part of JSON body.

o

However, isn't there a security risk here of sending password in plain text?

o

2.

o

For Sing-in I'm thinking about using CODETERM2 in Dropwizard.

o

3.

o

I don't want to store passwords in plain text.

o

What strategy should I follow after I get the users' password in the CODETERM3 as plain text?

o

I'm looking for some java libraries that can assist in password salt and MD5 .

o

Have you read their [authentication]( URL_http://dropwizard.codahale.com/manual/auth /#oauth2) documentation?

o

OAuth2 is going to be the most secure and RESTful approach; as for sending creds in plaintext, are you building a web front-end?

o

If so, serve your site over https and it won't be a problem.

o

I don't envision any external apps using the API for data.

o

My api will just be used by my app's mobile and web interface.

o

Wouldn't Oauth2 be an overdo for this?

o

OAuth2 is a great approach to ensure that each logged in user receives their own access token that you can track and manage.

p

Short of dropping the API altogether (and using just session/cookies) it's still the best and most scalable approach.

p

Do the work now, it'll last (forever?

o

).

o

Maybe I'm not following correctly.

p

If I go down OAuth2 route I have to register my app with one of the OAuth2 providers e.g.

o

google, github, etc.

o

right?

o

And then I can use that token to authorize the user?

o

No no no... Dropwizard supports OAuth2 on the back end - so you would implement your own OAuth2 service, and then have the browser perform the auth handshake with it; completely independent from Google, et al.

o

Ah, roger that.

o

much better.

o

I'll find more info on how to implement OAuth2 with drop wizard.

o

Users can "signup" with Oauth2 as well?

o

Yes they can.

o

Looks like their docs are pretty good too, and being open source, you should find lots of examples.

p

Looking at the docs, we can see that Dropwizard supports a standalone OAuth2 implementation: URL_http://dropwizard.codahale.com/manual/auth/#oauth2 [ URL_http://dropwizard.codahale.com/manual/auth /#oauth2] OAuth2 has several advantages, many of which can be read about here: URL_http://stackoverflow.com/questions/7561631/oauth-2-0-benefits-and-use- cases-why [OAuth-2.0:-Benefits-and-use-cases-why?]

p

Things to note: when dealing with authentication, you should always host over HTTPS to ensure transport encryption Dropwizard claims their OAuth2 implementation isn't yet finalized, and may change in the future.

o

As a fall back, they do support Basic auth as well, which when used over HTTPS would be still reasonably secure.

p

Implementing this does not involve using any third party "social" authentication services such as Google or Facebook.

o

Thanks for the shout out for the Dropwizard OpenID project.

o

Glad it was able to get you started.

p

If you want a pure web form type approach, take a look at another of my projects URL_https://github.com/gary-rowe/MultiBitMerchant [MultiBit-Merchant] which provides multiple authentication methods (web form, HMAC, cookie).

o

You'll need to dig around to really see it working since this project is not designed as a demo as such and is very much a work in progress.

p

After loading the project, look for URL_https://github.com/gary- rowe/MultiBitMerchant/blob/develop/mbm-client/src/main/java/org/multibit/mbm/c lient/interfaces/rest/auth/webform/WebFormClientAuthenticator.java [CODETERM1] which will get you in the right area.

o

The general principles involved with Dropwizard authentication URL_http ://gary-rowe.com/agilestack/2012/10/23/multibit-merchant-implementing-hmac- authentication-in-dropwizard [are-discussed-in-this-blog-article] .

o

Although it targets HMAC you can easily adapt it for web form or cookie using the source code referenced earlier.

o

It's all MIT license so just use it as you need.

o

Thanks.

o

I am looking through the code and implementing my simple service in parallel.

p

I had one more question - URL_http://stackoverflow.com/questions/20662871/how-to-do-basic-authentication - of-a-resource-in-dropwizard.

o

Open Source Iris Scanning Library in Java.

p

I am looking for suggestions about open source Iris Scanning libraries in Java.

p

Has anyone used before which can built an unique profile for each scan & be able to authenticate too.

o

Thank you.

p

I haven't used Java libraries that "just do" that entire work you describe, but it's not that difficult to use ( URL_http://sourceforge.net/projects/javavis/ [ URL_http://sourceforge.net/projects/javavis/ ] ) as in the sample code described in the following paper: URL_http://imai-software.com/openlab/data/IMAI- JOUNALVI.pdf#page=43 [ URL_http://imai-software.com/openlab/data/IMAI - JOUNALVI.pdf#page=43] (see from page 43) - that paper is quite good and for it's length :).

n

Of course, there are a few "working" solutions, e.g.

p

: URL_http://mac.softpedia.com/get/Math-Scientific/JIRRM.shtml [ URL_http://mac.softpedia.com/get/Math-Scientific/JIRRM.shtml ] (Java based) URL_http://projectiris.co.uk/ [ URL_http://projectiris.co.uk/ ] (QT based -but there are more QT <-> Java bridges) but personally I haven't used in production (just for testing) these 2 libraries since the documentation is not very good/detailed, and since this kind of problem is quite "cutting edge" so the documentation and an explanation paper is much more important than any source code.

p

If none of the above approaches work for you, I would suggest to use a MATLAB based iris recognition examples (because there are many MATLAB based implementations, papers, theses) and than convert them to Java or use the directly from Java (since MATLAB has support for this).

o

Link to the paper is broken.

n

I would look into URL_http://opencv.willowgarage.com/wiki/ [OpenCV-which- is-a-C++-library-for-computer-vision] .

o