09-07-2010 02:34 PM
@CMY: I think the "shared library" issue is definitely a huge one. Though I don't have any apps on App World I have (and am working on) general libraries that other devs can use in their own application. It is a pain to have to figure out how multiple apps can use the same library without it breaking one or the other.
Though I don't think there is a talk about this at Dev Con I hope someone brings it up to one of the RIM devs. Maybe they can remedy a solution to this.
09-07-2010 05:06 PM
Yeah I hope so, it's been mentioned on the forums plenty of times. I logged an issue on the issue tracker about the ApplicationPermissions, but have not heard anything on it (not surprising as I logged an issue over a year ago that only got one response after about two weeks, and none since).
The thing I don't get is it since it is being installed programatically, there has to be a way to specify more than one app being installed and install multiple jad entries. The fact that the shared library issue is so noticeable (esp on older OSes) makes me wonder why they refuse to look into it more. Even if the newer OSes can get rid of points 2 and 3, there is no way to get rid of point 1. Unless the library is installed sparately, there is no way to know the version number when installing to make sure you don't downgrade. And when RIM does so may OS versions, it would be improper for them to suggest developers finalize their shared resources, when they cannot do so themselves.
09-07-2010 07:03 PM
I wonder if RIM isn't ignoring it but just can't figure out a solution to it. I think that the OS keeps a fully named function on a giant stack in the OS.
This would explain a lot of the issues that exist (need to restart when apps are updated, why it takes so long to startup, why you can't have different libraries/apps with the same function in the same namespace, etc.) and would mean that if a shared library is loaded it would load the library's functions, and link them but if a separate app loads the same library it tries to reload and relink the functions, causing errors, the library not to load, and it to break. It would also mean that if the app that loaded the library was removed it would remove it's dependency and thus the apps that link to it will crash and say no classdef found.
It makes it near impossible to update unless the list's indexes don't change and the list is scanned for any old references to determine if the library should remain or be removed. Sounds easy but not if the whole OS is built on the idea that the indexes change when items are added or removed.
Just what I think might be happening.
09-07-2010 07:12 PM
At DevCon last year there was a round table about shared libraries, the general consensus was that they would be a good thing. Someone in RIM was tasked with looking into it.
I agree completely with rcmaniac25, and the model that is described (Big stack of loaded classes) ties in with how I thought it worked. In fact I believe you can have multiple classes, however only the first one will be loaded and used, the duplicate ones will be ignored with a "duplicate class" message.
09-07-2010 07:40 PM
Someone should ask again at this year's DevCon to see if anything has changed or if the "person assigned to the task" just kind of ignored it.
02-06-2011 07:53 AM
I have only developed for Blackberry devices in the last 8 months, I have three apps on app world.
One app in particular has been doing very well, but i have got two bad reviews in the past 3 months that has damaged my sales hugely.
These reviews are:
02-06-2011 10:29 AM
System needs fixing....agreed....
* No Developer can post a review...end of story or face suspension
* Vendors should be able to post small response to any review
* Removal of a review should result in star rating change
* Allow for reviews via PC - simply issue a transid that can be used to validate
02-06-2011 11:52 AM
@Peter - Denying these reviews is not the problem.
the problem comes in when the star rating stays affected by the bad review, irrespective of weather the review has been denied. The star rating on our apps is what make them sell. So denying the review does not make much of a difference if the star rating does not get restored.
If anyone can buy our app and just give a review without us responding or even contact support to assist with a problem (in which most cases it is not our apps fault) what stops competition from giving us "One man team small developers" bad reviews to eliminate competition.
These bad review i have received are unfounded and don't even contact me for support ?
If i where to buy a product and it does not work i wont just be like "O well there goes my money ...." i would fight to get my monies worth.
I don't mind bad reviews, if the sound is bad, if the graphics are dated if the functions don't work then give me a review by all means after i have had a chance to address your concerns.
One bad review can cost me $1000 + a year ?? i am sorry but that is not on ...
Please tell me if i am unreasonable.
02-06-2011 01:53 PM - edited 02-06-2011 02:04 PM
I would like to see two changes:
1) Disallowing negative reviews where the primary concern is that the app doesn't perform a certain function that the app was never advertised as having in the first place. For example I have an app that displays certain info from a web service. This is all the app is meant to do. However, every negative review that I've received is like this "App doesn't work.. it doesn't do X"
2) Being able to respond to reviews. It seems very one sided to me right now. On many websites such as tripadvisor.com one of the best things about the reviews is you can see how the vendor responds (or doesn't) to issues.
Also, I think the petition is abit flawed. There has to be more than two choices. It's like asking a question like this
Who should we invade?
That's abit extreme of an example but I hope you understand my point.
It's unfortunate that many only leave reviews when there's an issue. I would be very happy if there was a way to encourage more reviews from all users. Any feedback is good.