Preliminary V6 feature list overview & progress

[Ray]

1.posibility to edit title and description of searches from site that you have no control over ( i mean ability to change title and or description by you choice after indexing)

This is difficult to provide a practical usage for. You would not want to edit the title and description for each page individually, and you would certainly not want to repeat the process every time you re-index.

Perhaps a more helpful feature would be to accept a list of strings which would be stripped out of page titles and descriptions when they are indexed. Still might be a very long list.

2.ability to add relnofollow tag

Do you mean the rel="nofollow" attribute as described here and here?

As Zoom does not currently score pages based on linking (since it is mostly meaningless within a single site, and not helpful until a truely large collection of external sites are indexed), this would not apply in regards to its ranking effect.

As for telling Zoom Indexer to not follow or index the link, you can achieve this by using the ZOOMSTOPFOLLOW tag.

While you can't add this to sites outside of your control, I'm not sure how useful it would be to support rel="nofollow", since this tag is usually used on links posted by forum/blog visitors, etc. and they tend to be for links which point to an external site (not within the same site you are already indexing). As such, these links would already be excluded from indexing - since Zoom is not designed to index external sites unless the site is explicitly specified.

3.use category name as searched tag

If you mean that you would like to search for all pages belonging to a category (without any specific search word), then this should be possible in V6. This was something we added for the custom meta field search. More information will become available soon.

[sizbut]

Howzabout pairing your tags.

For example <!--ZOOMSTOP> and <!--/ZOOMSTOP-->. Much easier to remember than the existing mismatched combinations. I appreciate you would still have to support the existing formats so not much of a saving for you guys in programming but might reduce some of the support queries.

[Ian]

Features hopefully in V6

This is a list of features that will probably get into V6 assuming we get time to implement them and we still think there is demand for them.

• A more flexible template page (greater control of specific elements to be displayed and their location).
• Boosting of entire domains
• Tag cloud generator
• Keyword listing generator
• Category results summary (x results found for category A, y results for category B, etc)
• More advanced authentication methods (e.g. configure Zoom to POST login details), in addition to the existing methods like HTTP authentication.
• Improve cookie support (e.g. store cookies sent to spider)
• Return x results per domain sorting option. This can be useful if you are indexing dozens of different domains into a single index.
• Configurable maximum word length
• Improvements to exact phrase searches
• FastCGI support. This feature might speed up the CGI search times by 100% or 200%. But not many servers support FastCGI, and not many customers are asking for more speed in the CGI. So demand is low.
• Restructure of the low level index format to support higher capacity (something around 5 million to 10 million pages with the 64bit edition). This sounds great in theory, but in practice there aren't too many of our customers that have even reached the 100,000 pages mark. So demand isn't high for this.

We post more details and some screen shots of what is in development over the coming weeks.

I think that the two functions you mentioned;
1. Keyword listing generator
2. Category results summary (x results found for category A, y results for category B, etc)
Would greatly improve the user experience.

If we could set up a page which allowed users to invoke a search within preset category, that would be truly useful. Especially if the categories were available before the search were invoked.
Ian

[Ian]

synonyn lists

You currently allow for the inclusion of synonyms but they have to entered individually. The ability to import a .csv file would greatly speed things up.

Ian

[David]

1. Keyword listing generator

If you need a list of all keywords it can be found in the dictionary file. What did you plan on doing with the list of words if you have it?

The ability to import a .csv file would greatly speed things up

In the current version of the software you can already import synonyms. You must be using an old version.

set up a page which allowed users to invoke a search within preset category

You can pre-select a category by using a hidden field on a HTML form. See this old post and the FAQ for details.

[Vernon]

Date related weighting

I'd value being able to weight results according to a date value associated with the page. There's a post at thread 2588 on this. My comment there was:

We have a number of sites that have events, news items or books/CDs that have dates. At the moment we are manipulating the ZoomPageBoost value based on a date-related algorithm in order to push more recent news or publications towards the top of the list based on the same search word. Having another weighting facility that understands dates (by inserting a ZoomDate value or another Zoom meta tag - and NOT just the date of the page creation, which is meaningless in database-driven sites) would be very helpful indeed.

[chriscsanyi]

What is the time frame looking like on this new release? Also with the "Native ASP.NET search script option" will you be able to reconfigure how the categories are searched? Keep up the good work.

[David]

We have an alpha version available now. Which we are making available upon request to a few people who need specific features. It is not complete functionally and probably somewhat buggy.

So we are on track to release it 3rd qtr of 2008 (as stated in the initial post). My guess is a beta release late July and the final release Aug / Sept. But it will be released when it is finished. Rather than being released on a fixed date.

I can't really comment on your question about categories as I am not sure what you mean by reconfigure in relation to categories.

[maryjili]

Would you please post a list of new features the current alpha version has and a list of additional features that the final version will have? Thanks.

[Ray]

The alpha release has all the features listed in the "definitely in V6" post at the top of this thread, and some extra ones.

The "features hopefully in V6" post lists features that we hope to have in the final release.

[maryjili]

Thanks. What is the maximum capacity of the 64bit edition now? 5 million or 10 million pages or more? And how fast is the search speed?

[David]

The alpha release is 32bit only. But we have done the code for the indexer to be 64bit even if not yet available. At the moment we haven't done any capacity testing and further index structure changes will be needed in any case before we could get into 2M+ page range. Using 64bit only removes some bottle necks (like max RAM) but doesn't effect other issues.

We also haven't done much benchmark performance testing as yet. We are expecting fairly dramatic improvements in exact phrase searching. But maybe not much (if any) for single word searches. You should see these speed differences in the current Alpha release.

[Ray]

For several million pages, you should look at the MasterNode product. It would also have practical benefits in terms of managing the indexes (as it would divide up the total indexed content into smaller, more managable sets for re-indexing).

[Ray]

This thread is now closed.

We have created a new thread with a more updated feature list here.