1. It took me a long time to download the tool, install it, signup at Amazon.com and then create my first bucket. I had to read a lot of help text. One needs determination to finish the job and non-technical users will not be able to do all this. Having said that, the download & install part of bucket explorer was easy. If you can embed the Amazon's signup process within your UI, it will help non technical users.

2. I will be sharing the bucket with some colleagues to share documents. It will be good if everyone can create their own subfolder / sub buckets within a bucket.

3. I had to try 4 different names for my first bucket because all other names were already taken. Can you make the bucket name easier?

4. The Folder tree on the left is not really Windows Explorer style. something does not sound right with the tree's sorting.

5. Can you provide an option to save the Secret Key & Access Key somewhere online, so I will not have to remember it. Just give me a user id & password which is easier to remember?

6. I will not be able to use it from my public library because I cannot install your software there. Can you make a Web Version?

Overall its a good product for version 1. You guys have done a good job.

Andip

Thanks for your feedback.

1. In the current version of API, Amazon has not provided any way for us to create an account for you. The user must go directly to Amazon.com and signup for the service. We have tried to document it and tried to make it easier. Yot you need Amazon.com account, a developer account and you need to subscribe to the S3 service. The signup wizards can be different based on your current subscription level with Amazon. So for now, you will have to follow these steps. Its a ONE TIME process and takes less than 2 minutes if you read this document:
http://www.bucketexplorer.com/documenta ... s3-account

2. Good idea, we will add the capability to create virtual folders in buckets. Just to clarify, there is no concept of a "folder" in S3. When we show you something stored in folders, we create file names with slashes in them. But this all will be transparent to you once we add this feature.

3. That is a difficult one. We will have to read your mind to guess what do you want as bucket name and what is available. But we will work on this, we just don't want to use random characters and we want it to be easy to remember.

4. We will fix it.

5. We will implement this. I don't know how many people will use it, but we will do it. The whole concept with our product is that you are trusting Amazon.com with your credit card, your data and your user id & password. i.e. you don't have to share anything with us. But if that is what our users want, we will add that as an option.

6. I won't answer this one now, but trust me, you will be able to use it from public library in about 2 weeks. Drop by again in 2 weeks and see what we have to offer.


THANKS A LOT for the feedback. Feedback from users like you is the only way for us to make the product better.

Saurabh

andip wrote:
5. Can you provide an option to save the Secret Key & Access Key somewhere online, so I will not have to remember it. Just give me a user id & password which is easier to remember?


Andip

Your S3 keys can be retrieved from AWS anytime as by logging in at Amazon with your email and regular password. In the Web Services section if you hover the mouse over the "Your Web Services Account" button you will see "AWS Access Identifiers". You can also have a new secret key generated.

points 2, 3 & 4 above have been implemented in Sept Beta. We expect to release a stable version with these 3 features by October 15th:
http://www.bucketexplorer.com/forum/topic134.html

This is being submitted with the intent of helping your development and receiving a free copy of the software for the feed back per your website as of 10/5/2007 6:50 pm CT.

-------------------------------
1. My system:
-------------------------------
WinXP Pro SP 2 (current with updates as of 10/5/07. I use AVG Antivirus, Windows Defender (spyware), and Windows XP Firewall. I quickly downloaded your install file to an external drive and had zero problems installing to my local primary HDD.

-------------------------------
2. Things I liked:
-------------------------------
Easy install, logical helpful GUI layout. Your context menus are also intuitive (ie., I right click and see options that I thought I'd see). The uploading process is fast.

-------------------------------
3. Suggested Improvements:
-------------------------------
The progress meter for moving the files is fine for moving 1 file, but it does not give me a context for multiple files being transferred. For example, I uploaded a directory with 40-50 jpgs. While your message "File is uploading - <CURRENT FILE NAME>" is helpful, I would have liked to see how far along I am in the total upload of all the files.

Second, It would also be a bonus, albeit a small one, to have some sort of display that tells me how many files are in the directory being viewed.

-------------------------------
CONCLUSION
-------------------------------
I have had zero bugs or substantive issues with the software. It did everything I expected and needed it to do. Thank you for making it. I'll be blogging on Bucket Explore at nielsendigital dot com.

wjnielsen:

Thanks for the feedback. We expect to release the paid version sometimes next month. I will send you the license key instructions of the paid release, as soon as it is up on our web site.

Both of the features that you have asked for are in our ToDo list. They are both a little tricky for large number of files (where there are thousands of files in the source folder or in the target bucket). We don't want to slow down the upload process to first count the total number of files being uploaded, but we do understand that a user wants to see that number. So, we need to find an optimal solution without affecting the performance. It may take a couple of months to implement this "count" feature, but we will do it.

Thanks again for the feedback.

saurabh

Saurabh,

For what it's worth, I would prefer the speed that your product operates at presently than the count feature. It is not a huge bonus in my mind to have the count, certainly not at the expense of upload speed.

One other addition in terms of feed back would be the ability to rename files on the S3 server. For instance, I uploaded a file that has a filename that seems to have no problems in Windows, but does not link well from the S3 server (perhaps an issue on my Ubuntu box). It would be nice to change the filename and not have to delete and re-up the file from my local machine.

Again, kudos to you all, this is a nice app.

When I double click on a bucket and its contents are displayed below in what I am calling the "Bucket Content Pane" (BCP), no one file is selected by default. When I right click on a file and want to execute a command (e.g., get the public l ink) the app does not know which file I am intending (even though I right clicked on it and received the context menu) and so displays nothing. Perhaps, that is by design perhaps not. If you want to require a left click selection first, an error message should direct the user to do so, rather than just doing nothing.


_____
This is a series of posts because I am trying to figure out what the anti spam parameter is hitting on that prevents me from posting this as one post.

Now let's say I left clicked on the first file in a bucket, of course getting the public U RL is a breeze. However, when I right click on the last file all the files between the first and the last are selected (including the first and the last) and we run into the same problem if I try and get the public U RL l ink. The app does not seem to know which file I right clicked on to get the context menu (assuming it see all my files as selected there's an ambiguity); consequently it does nothing. If selection by right click is not an option for Bucket Explorer, an alert of some kind to let me know that I need to select only one file OR even better, not displaying those features that require only a single record id in the context menu.

Let me just caveat all this by saying that it was not difficult to figure out what I needed to do. However, I am a developer too so I think like a developer (hopefully). Your average user may get frustrated with the above issues mentioned.

Yours,
Will Nielsen

PS forgive the "U RL" business, but I could not post to the forum with the word without the space because of the anti-spam parameter you have in place.

wjnielsen wrote:Saurabh,
One other addition in terms of feed back would be the ability to rename files on the S3 server.

Rename is one of the most frequent feature requests at Amazon S3. We just hope that Amazon will support that some day. In the mean time, we are planning to add a button which will first download, then upload with the new name and then delete the original file.

wjnielsen wrote:the app does not know which file I am intending (even though I right clicked on it and received the context menu) and so displays nothing.

This was fixed in Spet Beta release (see the first sticky post for a link). I will test it again, and if this was not fixed then it will be fixed for sure in Oct 15th release.

wjnielsen wrote:when I right click on the last file all the files between the first and the last are selected (including the first and the last) and we run into the same problem if I try and get the public U RL l ink.

We fixed this in the most recent release by allowing to select multiple files and generate URLs for multiple files at the same time.