11 Anti-spam Products Tested

An anonymous reader writes "When we achieve world peace, that's when we'll get the perfect anti-spam solution. In the meantime, ZDNet has a comprehensive review of eleven of the latest anti-spam products including solutions from BitDefender, Clearswift, CA eTrust, GFI, IronPort, MailGuard, McAfee, MessageLabs, NetIQ, Network Box and Symantec Brightmail."

SpamBayes?

[opusman]

Seems like a glaring ommission.

Yawn - No OSS

[OnceWas]

Where are the OSS products? No Spamassassin?

Some review...

SpamAssassin

[AnotherFreakboy]

I wonder why they didn't mention SpamAssassin. Open Source solutions will never gain the market share they deserve if media never gives them the attention they deserve. And the media will never give them attention until they get market share. It's a deadly cycle. Note: Open Source does not inherently make a product worthy of market share.

SpamBayes...

[John Miles]

... unlike the other products reviewed, doesn't advertise on ZDNet.

Uhh

[pHatidic]

Why not just use thunderbird [mozilla.org], it already has pretty good anti-spam capabilities in it to begin with and it's free and open source. I will admit I only installed it a few hours ago so I haven't been using it very long. The reason I installed it was because Eudora for OS X was very slow and for some reason was deleting my newest email every time I tried to download new email. Thunderbird is extremely fast, has better features, no popups, and is free. So far I have encountered no bugs, except some of the spam filtering features were a little unintuitive so I had to try them all out to see what did what.

Spam "products"

[Anonymous Coward]

It seems that many people these days now just look to pick up a pretty box at the store to help deal with spam. However, anyone who does this must not get important e-mail. I, for one, don't want my e-mail being filtered by some proprietary application like mcafee with limited configurability and disclosing details on how it works for "trade secret/IP" reasons. If it's an ip blocking service, I want statistics and to know how IPs get on it. If it's something statistical, I want to know exactly what it does. It is very dangerous to let your correspondence get picked apart by a "black box."

Re:Yawn - No OSS

[OnceWas]

And yet - if true - this (and OSS) gets no mention in the review.

And nobody on the review team thought of this?

Re:Uhh

[alen]

this is for server side anti-spam products. some people want to stop spam before it gets into the email system

Re:SpamBayes?

[snorklewacker]

Er, the glaring omission would be any mention of the effectiveness of any of these products. Am I not clicking on the right links? Because I'm seeing less than a page worth of review for each product, that seems to consist of installing it, clicking around the admin interface, then going on to the next product. It doesn't appear that they actually used the products they were reviewing!

Re:SpamAssassin

[RealAlaskan]

Open Source solutions will never gain the market share they deserve if media never gives them the attention they deserve. And the media will never give them attention until they get market share.

How about:

``Open Source solutions will never gain the market share they deserve if media never gives them the attention they deserve. And the media will never give them attention until they [the Open Source solutions] start spending big bucks advertising with the media''.

No chicken-and-egg stuff here: I would bet that ZDNet is following thier long-standing policy of reward^H^H^H^Hviewing their advertisers' products.

ZDnet = paid promotion

[Anonymous Coward]

Notice there are no free products listed. If you "contribute" some ad revenue to ZDnet, they'll look at/write about your product. Otherwise....

Re:IM better than that

[eclectechie]

thats why its good antispam - cos i have to approve everyone on my list ;)

You can do that with email, too; block everything not explicitly whitelisted.

IM has no advantage here.

The test is stupid!

[Anonymous Coward]

This review does not actually test the ability of the software to catch spam. It is just a beauty contest.

The catch...

[Jayde Stargunner]

The other noteworthy point of the last page is the absolutely ridiculous statement of, "Please note that these decisions were not based on accuracy testing."

I'm sure everyone is just amazingly psyched about an "ultimate" anti-spam guide that makes no effort to determine if the products they are reviewing (let alone proclaiming as the "winner") actually stop spam.

Of course, I guess this kind of article is developed to benifit CIOs with no technical experience, who just want something to tell the IT department to install. (Thus: price and ease of installation are far more important than it actually doing what it is supposed to.)

It's a Beauty Pagent

[R_Harrold]

That is what this article is. It fails to address some of the most significant issues to be considered when selecing an anti-spam product: 1. What percentage of incoming SPAM does it catch? 2. What percentage of the messages caught were "Non-SPAM" messages? 3. What is the message volume the product can handle? Instead they gloss over catch rates and false positive rates with a "Everything does a similar job" type statement. FALSE. I just spent the past 8 months evaluating anti-spam solutions for my workplace and they are not all the same when it comes to spam catch rates. I don't really care how pretty it looks or how easy it is to install. Nor do I give a hoot about the buzzwords a particular product incorporates, give me the spam blocking accuracy and the volume it will handle. It is all fine and dandy to ignore volume when you are running a 200 user ISP, but when you get up to 50000 users with over a million messages a day it becomes slightly important. Robert H. Houston, TX

Testing?

[Grey]

This does look like a test to me. It look like a bunch of marketing speek.

We did not perform any "official" accuracy and performance testing on the products. We set the programs up in modes to test both controlled and live messages, however the results of these brief tests would just add more confusion to the mix than anything and certainly didn't show any unexpected results.

or the MS exchange 2003 only product got 2.5 stars and many others got 3?

Every product review is like, it installs easily, and quickly. So what, are you as sysadmin or moron?

A test should give performance facts like false negative and false positive rates. This is nothing but a bunch of marketting crap and should not be posted.

Use linux to protect exchange

[drsmack1]

Setting Up a Spam-Filtering Mail Gateway For Microsoft Exchange Using Fedora Core 1, Postfix 2.0.19, Amavisd-New and Razor2 http://tinyurl.com/3khzk