What is this site about? Homepage Our awards Rating criteria Key to our descriptions Software directory Winners Gallery What's new? A B C D E F G H I J K L M N O P Q R S T U V W X Y Z    More... Other operating systems Appendix Related software About us... Submit software Who are we? Link to this site! Terms of use		Rating Criteria The need for drill & practice software List of requirements           Aim of this survey This survey focuses on flashcard-based drill programs. Such programs are a type of CAL (Computer-Assisted Learning) software that ask you short questions and expect short answers. This scheme proves very effective for memorization purposes. Now let us tell crucial from minor features. To foster effective learning, a flashcard-based drill program must be selective: it has to be able to filter out long known questions and let you focus on the harder ones. This feature is discussed below in The need for drill & practice software. User-friendliness is also an issue, especially in the long run. Beware -- we do not mean 'ease of use' as it is understood by many. We think it is worth taking some quirks into the bargain, in case they are justified by a higher functionality. These and other features (slideshows, multimedia support, networking, scheduling, etc.) are discussed in the List of requirements below. Scope of this survey We are open to accepting various kinds of programs for review. But please note that we rate their fitness for learning purposes only, regardless of their actual aim. This means that programs destined for other purposes, such as quiz or test editors and administrators, will be evaluated with regard to features that may be secondary to them. Likewise, language- or subject-specific learning programs are rated solely with regard to their fitness for all-purpose memorization, and hence built-in limitations will affect the rating negatively. Please keep this in mind when interpreting a rating.

The need for drill & practice software

1. The need for repetition

There is a lot of evidence supporting the notion that repetition is necessary to prevent the effects of forgetting. This can be exemplified by so-called "forgetting curves", like the ones you can see in the picture. In our example, only one day after learning new material, up to 50% of that material is virtually forgotten (red curve), i.e. it cannot be brought to mind without help. But if you review the same material one day later, your memory trace will deepen, and hence the new forgetting curve will be less steep (green curve, first from the left). Additional repetitions may be spaced more and more, since the forgetting curve gets flatter with every session. Of course, the slope of a forgetting curve depends on many variables, but if you want to beat it, it is clear that you will have to schedule periodical reviews. Now, when trying to achieve this goal, you will be confronted with one obstacle: you cannot review all of your materials again and again. See the next section for a solution.

2. The need for flashcards

In certain English-speaking countries, so-called flash cards made of paper or cardboard are used in schools as a learning aid. Although flashcards are used mostly for reference purposes, you can think of a more sophisticated application than this: you can make each card bear a question and an answer (the latter should be written overleaf) and use them to test yourself by trying to recall the answers. Paper flashcards are often been used in this manner for learning vocabulary, historical dates, and mathematical formulas. But even further, since flashcards present the information fragmented, they give you a chance to establish priorities by gradually setting aside the flashcards you already know. That would prevent you from wasting time with long-known material. Let us explain a smart but simple procedure for this purpose.                                You know the right answer You can't recall the answer     In the early 1970s, the German psychologist Sebastian Leitner described a 'learning machine' (Lernmaschine, Lernkartei, Karteikasten) in a book he wrote for young students. Leitner's machine consists of five or more compartments (slots, pigeonholes). It works as follows: the user tests theirself on a flashcard contained in one of the pigeonholes. The flashcard is promoted to the next compartment if the question on it is answered correctly by the user, and, on the contrary, it is put back into the first compartment in case of failure. Thus, this "learning cardfile" gives positive feedback about the present state of the user's knowledge at any time: the lowest pigeonholes contain the hardest stuff, and the highest pigeonholes contain items that have been reviewed successfully several times. This overall view allows the user to set their own priorities without any major planning or scheduling efforts -- simply by choosing the appropriate pigeonhole. Another advantages of this method are that it is incorruptible and that learning is fairly stress-free. Now, you may argue that it is very awkward to cut out and label paper flashcards. You are absolutely right. That is why many developers have implemented this principle into computer programs. This website contains a collection of reviews of more than 200 "flashcard programs", most of which you can download from the Internet. In the next section, we will outline which features would make such a program an effective and powerful educational tool. You will see that memorization software can have further interesting advantages over paper flashcards.

List of requirements ( Press ALT+ CURSOR LEFT or CTRL + CURSOR LEFT  to get back to this table )

1. File formats

Let us call a collection of flashcards "a Pack of flashcards". A good flashcard trainer should allow edition of flashcard files. If an editor is not incorporated, the file format must be a standard in common use. 1.1 Flashcard style Almost any kind of question can be implemented in the plain question/answer (Q&A) format. Additional fields to the flashcards only contribute to make import/export procedures harder. These are some kinds of test styles: 1. Question & Answer Even questions including several answers that are acceptable can be implemented in Q&A format. 2. Yes/no questions They can also be implemented as Q&A. 3. Multiple-choice tests can be implemented in plain Q&A format as well, and this form is actually the most flexible bet, since no limit is set to the maximum number of options, etc. Anyway, flashcard makers' life is easier when automatic filling of multiple choice tests is available: the program displays randomly taken (wrong) answers beside the right one (written on the flashcard). (Of course, this would only make sense if the answers to all cards lie in the same conceptual category, e.g. history landmarks versus years. But what an advance!) Among others, the programs Fachwort 8.07 and Vocabulary Wizard 5.5 can do so. 4. Cloze tests: gap-filling or "fill in the blanks" tests can also be implemented in Q&A format. Flashcards could also contain hints (cribs, cues) the user could ask for in the training mode, and explanations to tricky multiple choice questions. Again, such hints should not consist in a separate field, for the reason explained above. In VTrain, such hints can be displayed or hidden at any time. 1.2 File formats within a flashcard: Whenever we mention "one-line text" in these pages, we will be referring to text that is not able to contain linefeeds or similar "control" characters, or to text that is displayed inside a field where only a single line fits into. Every question (or answer) should be able to contain plain text (.txt) or rich text (*.rtf, *.htm), a sound file (.wav, .mp3), an image file, or a video (".avi" or Quicktime®) file. Implementation by way of object linking and embedding (OLE) capabilities would make the flashcards much more powerful. By the way, mathematical and other scientifical formulas are handled best in MathML or TeX format. It is important to note that text formats other than plain text notably increase the difficulty of coding the procedure of checking the user responses with the right answers. (Workaround: the user would have to "grade" her/himself for each answer, using a keyboard shortcut.) (VTrain supports OLE insertion of any kind and number of objects on the same flashcard. WinFlash and Langenscheidts Vokabeltrainer support video files, but not embedded images in text files (OLE). Larry's Flash Cards is HTML-based. Only Flashcards 2000 by Capoeira includes an integrated, rudimentary graphics editor. Flashcard Pro supports OLE.) 1.3 Packs of flashcards: inner structure 1.3.1 Subcollections? In some flashcard-based drill programs, Packs of flashcards are grouped into collections and subcollections (sometimes called "lessons" and "units", etc.) They are intended to make it easier to narrow the goal range of each learning session. Our opinion is this complicates the inner structure of a Pack of flashcards unnecessarily. It is better to save Packs of flashcards in different folders. 1.3.2 Hierarchical lists A Windows-Explorer-like hierarchical list of flashcard files would also ease drag and drop of flashcards across collections, and help to overcome the limitations of the default Windows keyword-based search algorithm (usually F3). (Such a tree-like implementation is found in Lernkartei 2.0 by Archivista, Kiwi '97, Ro-Winvokabel and TommyB's Vokabeltrainer. MemorizeIt includes a non-hierarchical file browser.) Such an application should be able to keep the alphabetical order even in non-Latin alphabets and writing systems (Russian, Chinese). 1.3.3 Statistics across files: monitoring the learning progress If you use many different flashcard Study Files, you need to keep the overview over your personal performance as per each one of them, directly from the Open dialog box. Thus, an improved version of that dialog box would show the names training files (Study Files) along with the total number of flashcards in each one and bar diagrams displaying the relative number of cards in each level or compartment forming the Study File. So far, none of the applications tested include such statistical information in a file exploration window. 1.4 Packs of flashcards vs Study Files It would be useful to integrate two kinds of files into such a program : 1. Packs of flashcards would contain only the question and answer pairs, but not the score record of a user. 2. On the other hand, Study Files would consist of a collection of dynamic links to those Pack of flashcards files, plus the training logs. Thus, any correction in the Pack of flashcards would be of benefit to all users, each one with their own training file (Study File) ! A further advantage of this feature is that the Packs of flashcards themselves can be kept subject-specific, regardless of the personal needs or the temporary goals of each individual. Note: In order to comply with the Leitner system discussed above, Study Files must be integrated by compartments. VTrain, VocabWorks, Wissenstrainer, LingoMaxx, and BrainSport comply with this very important condition. 1.5 File size Each one of them should be able to contain an arbitrary number of cards, and the code be so efficient that putting a reasonable number of cards (1000+) into a single Pack of flashcards would not devour the system's resources on slow computers (e.g. 386). 1.6 File paths Location of files must be customizable. Being allowed to choose a work directory will spare the user having to move them into a default folder. 1.7 Sample files The purpose of this test series is to find all-purpose, flexible, and powerful flashcard learning software. But this does not mean that we will reject reviewing requests for vocabulary learning programs aimed at specific languages, since many of them can be used for languages other than those supplied as well. So, when we test an application, our focus will be on usability, not on nominal purpose. Good sample Packs of flashcards or vocabulary files available on a website will be considered a bonus, while they will not affect the final rating of the program.

2. Learning

2.1 Learning modes (See also 1.1 Flashcard Structure) 2.1.1 User input Q&A with an optional time limit (card marked as missed if not answered, answer may optionally be displayed). Typically, the user would type the answer in. Alternatively, they could just state if they know the answer. This could be done by shortcut (user's initiative) or after a prompt (just "Do you know the answer?" or Displaying the right answer + "Did you know the answer?"). 2.1.2 Slideshows The time limit feature can be used for a further "slideshow" mode: there, you engage in a "passive" exposure to a subject matter (good for getting into a new lesson). 2.2 Shortcuts (for the learning mode) Especially the testing mode should be as user-friendly as possible. Regardless of an individual's habits, this implies that the program should make available keyboard shortcuts to all functions, so that using the mouse is not a "must". This is of special importance to users with some disability. Make sure that the main responses are assigned key shortcuts, enabling extensive use of the numeric keypad, for example. Some suggestions for authors of flashcard-based drill programs: -o- The user states he or she knows the correct answer: SHIFT + Spacebar -o- The user states he or she does not know the answer: CTRL + Spacebar, or period (decimal point) on numeric keypad. -o- Skipping a question: Esc key or 'plus' sign on numeric keypad. -o- Cribbing (displaying the answer): see the right answer and still be able to mark the card as got right thereafter. For flashcard formats with several fields: -o- Choosing one from 5 multiple choice answers: keys 1-5 on numeric keypad or central keyboard. -o- Moving across forms or tables: TAB or number 8 on numeric keypad. -o- Same, but back: SHIFT + Tab, or number 7 on numeric keypad. 2.3 Assessment Selective learning is guaranteed only if the software sorts flashcards out in order to let you make rational use of your study time. For example, this can be achieved by way of Sebastian Leitner's five-stage flashcard method (see above). Anyway, Other methods can be implemented. In any case, the method in question must be flexible enough to allow the user to choose the level of difficulty to focus on. 2.4 Fault tolerance The check procedure of the program which is in charge of providing feedback to the text the user inputs should be able to tolerate certain typing mistakes or variants (omission of punctuation marks, blank spaces, capitals, resorting). Thus, it would allow several 'right' solutions without having to record them separately in the answers list. Customizable options for this feature would be useful. 2.5 Inverted questions You must be able to swap questions & answers, i.e. to ask the "answer" and expect the "question" to be input by the user. So, we could use a French-English vocabulary list, for example, as an English-French one. An alternating "random" option would also be interesting. 2.6 Statistics A bar diagram (etc.) showing the current number of cards in each of the Leitner compartments and the total number of cards of the Pack of flashcards in use should be available at any time. Statistical data about how many times a card was tested, how long each user spent with a file, etc. would also be interesting. 2.7 Scheduling The program must be able to schedule successive reviews throughout a school term (etc.). VTrain keeps the user informed about which flashcards are "due" for repetition, according to a spaced learning scheme (repeat after 1 week, 2 weeks, 4 weeks, etc.). In a similar manner, Langenscheidt's Vokabeltrainer sets up timetables created from user preferences. In Supermemo and A+ Learning scheduling does not only take place automatically, but it is integrated into the training feature. This means that the questions the user is asked by the program are selected according to a schedule. The drawback of this approach is, the process is not too transparent to the user, who is deprived of the power to choose what to learn when time is scarce. (Supermemo's approach is the most sophisticated one: schedulign is based on guesses it makes about the future development of the "forgetting curve" for a given subject, taking into consideration the past performance of the user.)

3. Edition

3.1 Flexible edition Standard, WYSIWYG typesetting ("What you see is what you get", no 'coding') is a must for a serious authoring system: free use of different fonts on the same flashcard, etc. Rich text (bold, italics, color, etc.) and insertion of images and videos are especially useful in certain environments. See File formats above for more information. Features such as voice synthesis and recognition are probably of little use for strict pedagogical purposes. To date, the first is tiring after some time, and existing voice recognition technology is not accurate enough to decide whether the user's pronunciation is good enough. 3.2 Multilingual support 3.2.1 Keyboard layout switch If you intend to type in two languages at a time, it can be quite awkward having to switch between two different (language-specific) keyboard layouts, even though Windows provides a keyboard shortcut for that. True multilingual support will make automatic keyboard switch available, detecting which language you are typing. VTrain takes an interesting approach: each of both sides of a flashcard is assigned a different keyboard layout, and the program switches the layout according to the cursor location. The same method was later implemented also by Open Book. 3.2.2 Character maps Using the Windows character map or typing ALT+000 sequences is no comfortable solution when you don't want or can't use a foreign keyboard layout. Customizable keyboard shortcuts are provided in VTrain for the characters contained in its internal character map. 3.3 Combining documents Drag&drop features would make Packs of flashcards management much more convenient, allowing us even to pick single cards and copy/move them across Packs of flashcards. Only VTrain provides a flexible interface that allows comfortable data exchange across flashcard lists (drag and drop, etc.) in a MDI (Multiple Document Interface). 3.4 Search algorithm Find a keyword in a Pack of flashcards or in a collection of Packs of flashcards. A replace option would also be interesting, but not necessary, if the Packs of flashcards can be exported/reimported. Many applications offer this facility in a more or less sophisticated manner. 3.5 Sorting algorithm Sort procedures should not simply follow Windows defaults, where all upper case letters (A...Z) come before all minuscules (a...z), and admit non-Roman alphabets. 3.6 Import / export facility If text files can be written in ASCII or ANSI, importation and exportation of wordlists and edition by way of powerful word processors is easier. Separators between fronts/backs and between flashcards should be customizable for these functions. Reminder, Teach2000.3, VTrain, and Langenscheidt's Vokabeltrainer have options flexible enough. 3.7 Printing People without a laptop computer :-) will find this feature useful: printing single flashcards. Many programs implement printing lists of flashcards, but not all print single flashcards: e.g. VTrain, Lernkartei 2.0 (the one by Archivista), Eurotrainer, Academic FlashCards.

4. Some advanced features

4.1 Multiple users In a school environment, a password input would be the condition to put the training scores into a user log file, which would be readable only with the user's and a Master password (teacher). Anyway, it would be useful for home users to bypass the password prompt at program start. 4.2 Networking capabilities The user logs (scores) file should lie on the school server, of course. A password protected specific module of the program would show the teacher the (cumulative) performance stats for each student in the class, also in form of a learning curve. See QuickCM and QuizPlease, for example. VTrain takes an approach which is rudimentary regarding network compatibility (there is no teacher module that collects user logs) but includes an interesting "smart update" feature that allows massive text replacement across files. 4.3 Installation Programs are usually distributed either by a zip file or a self-extracting installation program. In the first case, a readme.txt or file_id.diz description file is the least we can expect. In the other case, a short description of the purpose of the program must appear during installation (many programs display merely a license agreement). If the installation routine modifies any files outside the installation directory (typically Registry keys), an uninstall program or the removal facility by way of Control Panel must be available. If neither of these requirements are fulfilled, the help file should at least mention which Registry keys etc. have to be deleted by hand.

5. Costs

5.1 Shareware Shareware is often defined as "try-before-you-buy software". We would rather say that there is much more to it. Many (if not most) shareware programs have their origin in the needs of individuals with some programming knowledge. They do use their own product, and decide to share it with others. This makes a key difference from 'commercial' programs, whose functionality is often cut down for fear of frightening away the average user. Half the time the result is, the most sophisticated programs are shareware, while commercial software is hardly more than a beautiful shell. Moreover, the shareware phenomenon fosters creativity and innovation, due to the concurrence of different approaches to the same problem, and the probability of making your suggestions and requests heard is much higher when it comes to a program authored by an individual. 5.2 Shareware or freeware? Yes, many programs are freeware. Some of them are powerful, some of them are not. But you cannot expect anybody to work for you at no cost for years. The figures speak - most freeware projects are discontinued sooner or later. Take this into account when you come to a decision, especially if the program of your choice does not have an export function. Besides, considering the token prices of many shareware programs, the "Shareware or freeware?" question should not be an issue at all.